Three-coordinate bis(phosphinimino)methanide derivatives of 'open shell' [M(II)] (M = Mn, Fe, Co) transition metals

Article abstract of DOI:10.1016/j.jorganchem.2004.06.050

An alternative two-step synthesis of the previously reported bis(diphenylphosphinimino)methane [CH₂(Ph₂P = NC₆H₃ⁱPr₂-2.6)₂] involving bromine oxidation of dppm and metathesis wit

Full text of DOI:10.1016/j.jorganchem.2004.06.050

Journal of Organometallic Chemistry 689 (2004) 3163–3167
www.elsevier.com/locate/jorganchem
Three-coordinate bis(phosphinimino)methanide derivatives of
ꢀopen shellꢁ [M(II)] (M = Mn, Fe, Co) transition metals
a,
b
Michael S. Hill *, Peter B. Hitchcock
a
Department of Chemistry, Imperial College London, Exhibition Road, South Kensington, London SW7 2AZ, UK
b
School of Chemistry, University of Sussex, Falmer, Brighton BN1 9QJ, UK
Received 8 June 2004; accepted 25 June 2004
Available online 14 August 2004
Abstract
i
An alternative two-step synthesis of the previously reported bis(diphenylphosphinimino)methane [CH2(Ph2P = NC6H3 Pr2-2,6)2]
i
involving bromine oxidation of dppm and metathesis with 2,6-diisopropylaniline is described. [CH2(Ph2P‚NC6H3 Pr2-2,6)2] is
readily deprotonated by the transition metal silylamides [M{N(SiMe3)2}2] (M = Mn, Fe and Co) to provide a series of exclusively
i
three-coordinate derivatives [{CH(Ph2PNC6H3 Pr2-2,6)2M{N(SiMe3)2}] which have been characterised by elemental analysis, mag-
netic measurements and, in the case of the Fe derivative, X-ray structural analysis. Reactivity studies have shown that even such
sterically demanding bis(phosphinimino)methanide ligands are prone to protonolysis and reductive [P(V) to P(III)] degradation
which limits their utility as stable platforms for further M(II) derivitisation.
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Three-coordinate; Manganese; Iron; Cobalt; Amides
1. Introduction
cally and coordinatively unsaturated ꢀopen shellꢁ (i.e.
<10 d electrons) derivatives of the divalent first row
transition elements Mn, Fe and Co, some of which have
demonstrated a spectacular reaction chemistry [6–9].
The common entry point for much of this chemistry
has been the synthesis of low-coordinate halide and
amido complexes. Our initial attempts to synthesise anal-
ogous derivatives stabilised by I indeed resulted in the
isolation of three- and four-coordinate compounds with
similar stability and electronic properties. Furthermore,
we reasoned that studies of the reaction chemistry of
three-coordinate amides, such as the amido-derivatives
III would be equally fruitful [3b]. We have noted that
many of our compounds feature a pronounced folding
of the ligand at the two four-coordinate geminal phos-
phorus(V) atoms [1–4]. This ligand non-planarity is in
marked contrast to the vast majority of b-diketiminato
compounds where planarity is effectively maintained
through charge delocalisation over the exclusively sp2-
hybridised atoms of the N–C–C–C–N chelate. This
We have recently reported synthetic and preliminary
reactivity studies of a variety of s-, p-, d- and f-block
derivatives supported by the sterically demanding
bis(phosphinimino)methanide anion (I) [1–4]. In these
compounds, a level of kinetic stability is attained due
to the steric demands of the N-mesityl substituents and
the Ph2P-ligand ꢀbackboneꢁ. Our original interest in this
class of ligand arose from the topological similarity of I
to N-aryl substituted b-diketiminato anions exemplified
by the 2,6-diisopropylphenyl (Dipp)-substituted deriva-
tive, II. b-diketiminates have attained recent prominence
as stable and robust supporting ligands for an enormous
variety of metal fragments [5]. Use of the anion II, for
example, has enabled the synthesis of several electroni-
*
Corresponding author. Tel.: +44-2-07-594-5709; fax: +44-2-07-
594-5804.
E-mail address: mike.hill@imperial.ac.uk (M.S. Hill).
0022-328X/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.06.050

3164
M.S. Hill, P.B. Hitchcock / Journal of Organometallic Chemistry 689 (2004) 3163–3167
feature of bis(phosphinimino)methanide coordination
chemistry results in a degree of interaction between the
central P–C–P methanide carbon and the metal centre
that is apparently subtly dependent upon both the over-
all metal coordination number and co-ligand identity
[3,10]. This carbanionic character indicates that anions
such as I are perhaps more realistically compared to
other heteroatom substituted methyl ligands,
(X)nH3 À nCÀ (e.g. X = R3Si, n = 1–3; X = R2P(III),
R2P(V); n = 1,2), resulting in a chelating ligand with en-
hanced basicity and reactivity. We hoped however that
the steric demands of the N- and P-bound substituents
would provide sufficient kinetic stabilisation to enable
the synthesis of a range of low-coordinate derivatives
in which the chelated ligand would behave as an inert
spectator in subsequent reactions. This was indeed the
case in reaction with relatively bulky acids where, for
example, protonolysis of the Fe(II) amide (IIIb), with
Ph3COH provided the triphenylmethoxy-derivative IV
[b]. All attempts to react IIIa–c with less sterically
demanding O- and N-acids were frustrated by competi-
tive protonolysis of the bis(phosphinimino)methanide
ligand I or, in reactions with reducing agents, reductive
degradation of the ligand P(V) centres [11]. We have
made similar observations with regard to the reactivity
of closely related three-coordinate zinc derivatives [3a].
In an attempt to enhance the kinetic stability of the com-
plexes, we turned our attention to the bulkier Dipp
N-substituted analogue of I, the anion V, which may
be obtained by deprotonation of the neutral bis(phos-
phinimino)methane, [CH2(Ph2P‚NDipp)2] (VI).
equivalents of 2,6-diisopropylphenyl azide [12]. As the
purification of this azide entails a potentially hazardous
distillation, this compound was used as a crude red oil.
In our hands, this protocol resulted in only poor yields
of the target bis(phosphinimino)methane. We sought
therefore an alternative route to VI (Scheme 1). Reac-
tion of two molar equivalents of molecular bromine with
dppm in CH2Cl2 produced a pale yellow solution and,
after removal of volatiles, a colourless solid that was as-
sumed to be [CH2(Ph2PBr2)2]. This was employed with-
out characterisation or further purification in reaction
with two equivalents of DippNH2 and four equivalents
of Et3N in toluene. Heating of this mixture to reflux
gave a colourless suspension from which crude VI was
isolated by filtration and removal of volatiles. Crystalli-
sation from diethyl ether produced a monosolvated
form of VI as large colourless crystals.
Addition of one equivalent of VI to a toluene solu-
tion of [M{N(SiMe3)2}2] at room temperature pro-
duced, after removal of volatiles, the monoamides
[CH(Ph2PNDipp)2MN(SiMe3)2] as colourless (1,
M = Mn; 2, M = Fe) and green (3, M = Co) solids.
Recrystallisation from toluene at room temperature pro-
vided the analytically pure crystalline compounds con-
taining a single toluene molecule of crystallisation. In
the case of the Fe(II) derivative, 2, these were suitable
for analysis by a single crystal X-ray study. Although
crystals of 1 and 3 appeared to be equally well-formed,
our attempts to obtain X-ray structural data were ham-
pered by weak diffraction at high angle. Solution mag-
netic measurements (Evansꢁ method) in C6D6,
however, gave leff(C6D6) values for all three compounds
that were consistent with high spin d5 (1, lB = 5.90,
S = 5/2), d6 (2, lB = 5.15, S = 2) and d7 (3, lB = 4.80,
S = 3/2) electronic configurations with five, four and
three unpaired electrons, respectively. Solid-state mag-
netic studies of the crystalline sample employed in the
X-ray study of the iron derivative 2 were also carried
out. A plot of 1/v(m) versus temperature in the range
5–300 K (Fig. 1) displayed simple Curie behaviour and
gave a leff of 5.16 lB. The similarity of this value to that
of the solution measurement provides evidence that a
similar three-coordinate structure to that determined
by the X-ray study (vide infra) also persists in non-coor-
dinating solvents, while the close correspondence of the
values obtained here to those of amide derivatives sup-
ported by the less sterically demanding ligand I and clo-
sely related b-diketiminato derivatives indicates that all
three compounds are, most likely, exclusively three-co-
ordinate in both solid and solution states [3b,6,7].
The molecular structure of 2 is illustrated in Fig. 2
while selected bond lengths and angles are listed in the
figure caption. The complex crystallises as the monomer
along with a disordered molecule of toluene solvent.
The iron is bound to one bis(phosphinimino)metha-
nide and a single N(SiMe3)2 ligand. As was the case in
Pri
Pri
R1 Ph2P
N
PPh2 R1
N
-
-
N
N
Pri
Pri
R1
R2
Ph
R2
R1
I:R 1 =R 2 =C H3
V:R 1 = iPr; R2 =H
II
X
Ph
Ph
M
P
P
Ph
N
Ar
N
Ar
Ar =2 ,4,6-Me3C6H2
III:M =a , Mn; b, Fe; c, Co.: X = N(SiMe3)2
IV: M= Fe; X = OCPh3
2. Results and discussion
The synthesis of compound VI has previously been
reported by Bochmann and coworkers via the oxidation
of bis(diphenylphosphino)methane (dppm) with two

M.S. Hill, P.B. Hitchcock / Journal of Organometallic Chemistry 689 (2004) 3163–3167
3165
Ph2P
PPh2
N
4 Et3N, 2 2,6-iPr2C6H3NH2
Pri
Pri
2 Br2
N
Br2Ph2P
PPh2Br2
Ph2P
PPh2
toluene,
CH2Cl2
- 4 Et3NH.Br
Pri
Pri
VI
Scheme 1.
100
80
60
40
20
0
three-coordinate. The N(1)–N(2)–N(3)–Fe coordination
plane reveals no significant pyramidalisation at iron
y= 0.3032x - 2.037
R2 = 0.9991
P
(
lengths in 2 [2.035(3), 2.034(3) A], although slightly elon-
angles = 359.93°). The Fe–N(1) and Fe–N(2) bond
˚
gated in comparison to those of IIIb [1.992(3), 1.997(3)
˚
A], are shorter than in the closely related four-coordinate
ꢀateꢁ complex [{CH(Ph2PNC6H2Me3-2,4,6)2}Fe(a-Cl)2-
Li(THF)(OEt2)] [3b]. There is a slight increase of ligand
bite angle in 2 [112.3(13)°] compared to that of IIIb
[110.87(12)°]. Although possibly a result of the increased
steric demands of the Dipp substituents, this value is
some 10–15° wider than the corresponding bite angles
of three-coordinate b-diketiminate derivatives contain-
ing the same metal and coligand [6–9]. It is more likely
therefore that this bite angle is influenced more directly
by the chemical constitution, and hence conformation of
the N–P–C–P–N chelate (vide infra). Consequently, the
remaining interligand angles around the iron centre are
more acute than those of IIIb [b]. The M–N(3) distance
0
100
200
300
400
-20
T/K
1/Chi(m)
Linear (1/Chi(m))
Linear (1/Chi(m))
Fig. 1. v(m) vs. T in the range 5–300 K for compound 2.
˚
to the silylamido substituent [1.931(3) A] is essentially
˚
identical to that of IIIb [1.932(3) A] and is within the
range previously observed in three-coordinate iron
derivatives that are terminally bonded to the N(SiMe3)2
ligand [13–15]. As we have noted previously [1–4], the
ligand framework of 2 provides evidence for considera-
ble delocalisation of charge via the P–N linkages. The
P–C(1) bond lengths are shortened in comparison to
typical P–C r bonds while the P–N bond distances are
elongated compared to the corresponding values of
structurally characterised bis(phosphinimino)methanes
[e.g. for [H2C(Ph2P‚NSiMe3)2] P–C(1) = 1.825(1); P–
˚
N = 1.536(2) A] [16]. The most notable feature of the
Fe–N–P–C–P–N metallacycle of 2 is the irregular ꢀflat-
tened chairꢁ conformation adopted. This results in a
transoid relationship of the N-bonded 2,6-diisopropyl-
phenyl substituents (Fig. 3) and contrasts to the pro-
nounced boat conformation observed in the structure
of IIIb. In this latter case, the less sterically demanding
mesityl groups adopt a cisoid configuration that is more
typical of derivatives of this ligand.
As outlined above, b-diketiminato derivatives of the
first row transition elements have given rise to a great
variety of unusual and striking reaction chemistry [5–
9]. The heteroleptic compounds 1–3 are rare examples
of stable of open shell (i.e. <10 d electrons) three-coor-
dinate transition metal complexes. Such low coordinate
centres are often implicated as transients in a wide vari-
Fig. 2. The molecular structure of 2 (30% probability ellipsoids). H
˚
atoms removed for clarity. Selected bond lengths (A) and angles (°):
Fe–N(1) 2.035(3); Fe–N(2) 2.034(3); Fe–N(3) 1.931(3); P(1)–N(1)
1.639(3); 1.640(3); P(1)–C(1) 1.714(4); N(1)–Fe–N(2) 112.33(13); N(2)–
Fe–N(3) 125.74(14); N(1)–Fe–N(3) 121.86(14); Fe–N(1)–P(1)
108.36(17); Fe–N(1)–P(1) 117.89(17); N(1)–P(1)–C(1) 114.31(18);
N(2)–P(2)–C(1) 110.47(18); P(1)–C(1)–P(2) 126.6(2).
the previously reported Fe(II) derivative (IIIb), there is
no contact between the methanide carbon centre and
˚
the metal [Fe–C(1), 3.279 A] which is unambiguously

3166
M.S. Hill, P.B. Hitchcock / Journal of Organometallic Chemistry 689 (2004) 3163–3167
Fig. 3. The N–P–C–P–N–M cores of (a) 2, (b) IIIb [b] viewed along the Fe–N(3) bond, emphasising the chelate conformations and the respective
orientations of the N-aryl substituents. For 2 the angles defined by the P(1)–C(1)–P(2) and N(1)–Fe–N(2) least square planes and the P(1)–N(1)–P(2)–
N(2) plane are; 36.74° and 14.27°.
ety of transition metal-catalysed processes such as olefin
polymerisation, C–H activation and the activation of
otherwise inert small molecules [17]. Based upon the re-
cent precedent provided by known b-diketiminato
chemistry, we were hopeful that reactions of compounds
1–3 would provide a route to novel chemistry. Accord-
ingly we have attempted reactions with a variety of
protic and reducible substrates. Addition of 1,2-diph-
enylhydrazine to solutions of 1, 2 or 3 produced
indiscriminate protonolysis of both the bis(phosphini-
mino)methyl and hexamethyldisilazido ligand, while
reactions with azobenzene gave no reaction at room
temperature and apparent decomposition, evidenced
from 31P NMR monitioring of the reaction in C6D6,
upon heating to 60 °C in toluene.
dppm (3.0 g, 7.81 mmol) in CH2Cl2 (60 ml). Over the
course of 1 h the initial brown colouration faded to give
a pale yellow solution. Removal of volatiles gave a
cream-coloured solid which was then suspended in tolu-
ene (80 ml). Triethylamine (3.40 g, 33.7 mmol) and 2,6-
diisopropylphenylaniline were added via syringe and the
resulting suspension heated to reflux for 14 h. After fil-
tration, removal of volatiles produced a yellow glass.
This was crystallised from diethyl ether to produce large
colourless blocks of VI as an ether solvate (4.36 g, 69%)
1
with H NMR data identical to those quoted in the lit-
erature [12].
i
3.3. [{CH(Ph2PNC6H2 Pr2-2,6)2}MN(SiMe3)2] 1,
(M = Mn); 2, (M = Fe); 3 (M = Co)
In summary, we have prepared an isoleptic series of
three-coordinate and divalent Mn, Fe and Co amides
supported by a highly sterically demanding bis(diphe-
nylphosphinimino)methanide. Despite steric bulk which
is comparable to a variety of b-diketiminato derivatives,
the ꢀcarbanionicꢁ nature and presence of reducible P(V)
centres appears to limit the potential of this class of lig-
and in subsequent reactivity studies.
These compounds were prepared by the same general
method. A solution of VI in toluene was added at room
temperature to an equimolar solution of the appropriate
silylamide, [M{N(SiMe3)2}2] in toluene. In the case of
the iron derivative this produced an immediate darken-
ing of the solution and a colour change from green to
brown. The solutions were stirred for 12 h at which
point they were concentrated to incipient crystallisation.
1, 2 and 3 were crystallised as mono-toluene solvates by
slow cooling of this solution from 60 °C and isolated as
large colourless (1, 2) or green (3) crystals in ca. 60–70%
yields. 1: Anal. Calc. for C62H81N3P2Si2Mn: C, 71.50;
H, 7.78; N, 4.04. Found: C, 71.10; H, 7.21; N, 3.85%.
2: Anal. Calc. for C62H81N3P2Si2Fe: C, 71.44; H, 7.85;
N, 4.03. Found: C, 71.45; H, 7.74; N, 3.95%. 3: Anal.
Calc. for C62H81N3P2Si2Co: C, 71.23; H, 7.82; N, 4.02.
Found: C, 71.10; H, 7.69; N, 3.91%. All analyses calcu-
lated as mono-toluene solvate.
3. Experimental
3.1. General considerations
All reactions were conducted under an atmosphere of
dry argon and manipulated either on a double manifold
vacuum line or in a dinitrogen–filled drybox operating
at 1 ppm of O2. Toluene was purified by distillation
from molten sodium. [M{N(SiMe3)2}]2 (M = Mn, Fe,
Co) [18–20], were synthesised by the literature proce-
dures. Elemental analyses were performed by SACS at
the London Metropolitan University. Solid-state mag-
netic meausurements were performed on a Quantum
Design MPMS5 SQUID magnetometer.
3.4. Crystallography
Data for 2 were recorded on a Kappa CCD diffrac-
˚
tometer by use of Mo Ka radiation (k = 0.71073 A).
i
3.2. [CH2(Ph2PNC6H2 Pr2-2,6)2] (VI)
The structure was solved by direct methods (SHELXS
-
97) [21] and refined by full-matrix least-squares on all
data (SHELXL-97) [22]. Non-H atoms were anisotropic
and H atoms were included in riding mode. The toluene
A solution of Br2 (2.50 g, 15.6 mmol) in CH2Cl2 (30
ml) was added dropwise at 0 °C to a stirred solution of

M.S. Hill, P.B. Hitchcock / Journal of Organometallic Chemistry 689 (2004) 3163–3167
3167
[5] L. Bourget-Merle, M.F. Lappert, J.R. Severn, Chem. Rev. 102
(2002) 3031.
solvent molecule was disordered. Chemical formula
C55H73FeN3P2 Æ (C7H8). Formula weight 1042.27; T
[6] A. Panda, M. Stender, R.J. Wright, M.M. Olmstead, P. Klavins,
P.P. Power, Inorg. Chem. 41 (2002) 3909.
ꢀ
˚
(K) 173(2); triclinic, P1 (No. 2); a = 12.2460(2) A,
˚
˚
b = 13.3237(2) A, c = 18.9399(3) A; a = 71.032(1)°,
b = 86.286(1)°, c = 84.644(1)°; Z = 2; V = 2907.77(8)
[7] (a) J.M. Smith, R.J. Lachiotte, P.L. Holland, Chem. Commun.
(2001) 1542;
A ; Dcalc = 1.19 Mg mÀ3
;
l = 0.40 mmÀ1
;
h
3
˚
(b) J.M. Smith, R.J. Lachiotte, K.A. Pittard, T.R. Cundari, G.
Lucat-Rodgers, K.R. Rodgers, P.L. Holland, J. Am. Chem. Soc.
123 (2001) 9222;
range = 3.75–25.08°; R1; wR2 [I > 2r(I)] = 0.059; 0.150;
R1; wR2 all data = 0.082; 0.164; measured/independent
reflections/Rint = 30984/10101/0.052; reflections with
I > 2r(I) = 7714.
(c) H. Andres, E.L. Bominaar, J.M. Smith, N.A. Eckert, P.L.
Holland, E. Munck, J. Am. Chem. Soc. 124 (2002) 3012;
¨
(d) J.M. Smith, R.J. Lachiotte, P.L. Holland, Organometallics 21
(2002) 4808;
(e) P.L. Holland, T.R. Cundari, L.L. Perez, N.A. Eckert, R.J.
Lachiotte, J. Am. Chem. Soc. 124 (2002) 14416;
(f) J.M. Smith, R.J. Lachiotte, P.L. Holland, J. Am. Chem. Soc.
125 (2003) 15752;
Acknowledgements
We thank the Royal Society for a University Re-
search Fellowship (M.S.H.) and Mr. Kieran Brigatti
for assistance with the SQUID meausurement.
(g) J. Vela, S. Stoian, C.J. Flaschenriem, E. Munck, P.L. Holland,
¨
J. Am. Chem. Soc. 126 (2004) 4522;
(h) T.J.J. Sciarone, A. Meetsma, B. Hessen, J.H. Teuben, Chem.
Commun. (2002) 1580.
[8] V.C. Gibson, E.L. Marshall, D. Navarro-Llobet, A.J.P. White,
D.J. Williams, J. Chem. Soc., Dalton Trans. (2002) 4321.
[9] X. Dai, P. Kapoor, T.J. Warren, J. Am. Chem. Soc. 126 (2004)
4798.
Appendix A. Supplementary material
[10] C. Bibal, M. Pink, Y.D. Smurnyy, J. Tomaszewski, K.G. Caulton,
J. Am. Chem. Soc. 126 (2004) 2312.
Crystallographic data for the structural analysis of 2
have been deposited with the Cambridge Crystallo-
graphic Data Centre, CCDC No. 240802. Copies of this
information may be obtained free of charge from The
Director, CCDC, 12 Union Road, Cambridge CB2
1EZ, UK (fax: +44-1223-336033; e-mail: deposit@ccdc.
cam.ac.ukor www: http://www.ccdc.cam.ac.uk). Supple-
mentary data associated with this article can be found,
in the online version, at doi:10.1016/j.jorganchem.
2004.06.050.
[11] M.S. Hill, S.M.A. Karagouni, unpublished results.
[12] S. Al-Benna, M.J. Sarsfield, M. Thornton-Pett, D.L. Ormsby,
`
P.J. Maddox, P. Bres, M. Bochmann, J. Chem. Soc., Dalton
Trans. (2000) 4247.
[13] D.C. Bradley, M.B. Hursthouse, K.M.A. Malik, R. Moseler,
Transition Met. Chem. 3 (1978) 353.
[14] (a) P.P. Power, B.D. Murray, Inorg. Chem. 23 (1984) 4584;
(b) U. Siemeling, U. Vorfeld, B. Neumann, H.-G. Stammler,
Inorg. Chem. 39 (2000) 5159 and references therein.
[15] M.M. Olmstead, P.P. Power, S.C. Shoner, Inorg. Chem. 30 (1991)
2547.
[16] A. Muller, M. Mo¨hlen, B. Neumuller, N. Faza, W. Massa, K.
¨ ¨
Dehnicke, Z. Anorg. Allg. Chem. 625 (1999) 1748.
[17] (a) C.C. Cummins, Prog. Inorg. Chem. 47 (1998) 685;
(b) S. Alvarez, Coord. Chem. Rev. 193 (1999) 13.
References
[1] M.S. Hill, P.B. Hitchcock, Chem. Commun. (2003) 1758.
[2] M.S. Hill, P.B. Hitchcock, S.M.A. Karagouni, J. Organomet.
Chem. 689 (2004) 722.
[18] H. Burger, U. Wannagat, Monatsch. Chem. 94 (1963) 1007.
¨
[19] H. Burger, U. Wannagat, Monatsch. Chem. 95 (1964) 1099.
¨
[20] R.A. Andersen, K. Faegri, J.C. Green, A. Haarland, M.F.
Lappert, W.-P. Leung, K. Rypdal, Inorg. Chem. 27 (1988) 1782.
[21] G.M. Sheldrick, SHELXS-97: Program for the Solution of Crystal
[3] (a) M.S. Hill, P.B. Hitchcock, J. Chem. Soc., Dalton Trans.
(2002) 4694;
(b) D.J. Evans, M.S. Hill, P.B. Hitchcock, J. Chem. Soc., Dalton
Trans. (2003) 570.
¨ ¨
Structures, University of Gottingen, Gottingen, Germany, 1997.
[22] G.M. Sheldrick, SHELXL-97: Program for the Refinement of
Crystal Structures, University of Gottingen, Gottingen, Germany,
1997.
[4] M.S. Hill, P.B. Hitchcock, J. Chem. Soc., Dalton Trans. (2003)
4570.
¨
¨