Inorganic-organic hybrids of the p,p′-diphenylmethylenediphosphinate, pcp2-. Synthesis, characterization, and XRPD structures of [Sn(pcp)] and [Cu(pcp)]

Article abstract of DOI:10.1021/ic050683p

Two new inorganic-organic polymeric hybrids [Sn(pcp)] and [Cu(pcp)], pcp = CH₂(PhPO₂)₂^2-, have been synthesized and structurally chracterized. The tin derivative has been obtained by reaction of the p,p′-diphenylmethylenediphosphinic acid (H₂pcp) in water with SnCl₂·2H₂O, while the copper derivative has been synthesized through a hydrothermal reaction from the same H ₂pcp acid and Cu(O₂CMe)₂·H₂O. The structures of these compounds have been solved ab initio by X-ray powder diffraction (XRPD) data. [Sn(pcp)] has a ladder-like polymeric structure, with tin(II) centers bridged by diphenylmethylenediphosphinate ligands, and alternating six- and eight-membered rings. The hemilectic coordination around the metal shows the tin(II) lone pair to be operative, resulting in significant interaction mainly with a C-C bond of one phenyl ring. The [Cu(pcp)] complex displays a polymeric columnar structure formed by two intersecting sinusoidal ribbons of copper(II) ions bridged by the bifunctional phosphinate ligands. The intersections of the ribbons are made of dimeric units of pentacoordinated copper ions. Crystal data for [Sn(pcp)]: monoclinic, space group P2₁/c, a = 11.2851(1), b = 15.4495(6), c = 8.6830(1) A, β = 107.546(1)°, V = 1443.44(9) A, Z = 4. Crystal data for [Cu(pcp)]: triclinic, space group P1, a = 10.7126(4), b = 13.0719(4), c = 4.9272(3) A, α = 92.067(5), β = 95.902(7), γ = 87.847(4)°, V = 685.47(7), Z = 2. The tin compound has been characterized by ¹¹⁹Sn MAS NMR (magic-angle spinning NMR), revealing asymmetry in the valence electron cloud about tin. Low-temperature magnetic measurements of the copper compound have indicated the presence of weak antiferromagnetic interactions below 50 K.

Full text of DOI:10.1021/ic050683p

Inorg. Chem. 2005, 44, 9416−9423
Inorganic
−
Organic Hybrids of the p,p
′
-Diphenylmethylenediphosphinate,
-
pcp2 . Synthesis, Characterization, and XRPD Structures of [Sn(pcp)]
and [Cu(pcp)]
Jens Beckmann,†,‡ Ferdinando Costantino,§ Dainis Dakternieks,† Andrew Duthie,† Andrea Ienco,|
Stefano Midollini,| Cassandra Mitchell,† Annabella Orlandini,*,| and Lorenzo Sorace
Istituto di Chimica dei Composti Organometallici, ICCOM, CNR, Via Madonna del Piano,
50019 Sesto Fiorentino, Firenze, Italy, Dipartimento di Chimica, UniVersita` di Perugia,
Via Elce di Sotto 8, 06123 Perugia, Italy, Centre for Chiral and Molecular Technologies, Deakin
UniVersity, Geelong, Victoria 3217, Australia, and Dipartimento di Chimica and UdR INSTM,
UniVersita` di Firenze, Via della Lastruccia 3, 50019 Sesto Fiorentino, Firenze, Italy
Received May 3, 2005
2
Two new inorganic−organic polymeric hybrids [Sn(pcp)] and [Cu(pcp)], pcp
)
CH2(PhPO2)2 -, have been synthesized
and structurally chracterized. The tin derivative has been obtained by reaction of the p,p
diphosphinic acid (H2pcp) in water with SnCl2 2H2O, while the copper derivative has been synthesized through a
hydrothermal reaction from the same H2pcp acid and Cu(O2CMe)2 H2O. The structures of these compounds have
′-diphenylmethylene-
‚
‚
been solved “ab initio” by X-ray powder diffraction (XRPD) data. [Sn(pcp)] has a ladder-like polymeric structure,
with tin(II) centers bridged by diphenylmethylenediphosphinate ligands, and alternating six- and eight-membered
rings. The hemilectic coordination around the metal shows the tin(II) lone pair to be operative, resulting in significant
interaction mainly with a C−C bond of one phenyl ring. The [Cu(pcp)] complex displays a polymeric columnar
structure formed by two intersecting sinusoidal ribbons of copper(II) ions bridged by the bifunctional phosphinate
ligands. The intersections of the ribbons are made of dimeric units of pentacoordinated copper ions. Crystal data
for [Sn(pcp)]: monoclinic, space group P21/c, a
)
11.2851(1), b
4. Crystal data for [Cu(pcp)]: triclinic, space group P1
4.9272(3) Å, R ) 92.067(5), â ) 95.902(7), γ ) 87.847(4) , V 685.47(7), Z )
)
15.4495(6), c
)
)
8.6830(1) Å, â ) 107.546(1)
10.7126(4), b 13.0719(4),
2. The tin compound has
°,
V
c
)
)
1443.44(9) Å, Z
)
h
, a
)
°
)
been characterized by 119Sn MAS NMR (magic-angle spinning NMR), revealing asymmetry in the valence electron
cloud about tin. Low-temperature magnetic measurements of the copper compound have indicated the presence
of weak antiferromagnetic interactions below 50 K.
Introduction
optics, and magnetism).1,2 Whereas a plethora of metal
phosphonates, with a variety of structural arrangements and
dimensionality, have been reported (see examples in refs
3-8), the number of metal phosphinates described is limited
(see examples in refs 9 and 10). This is most likely due to
Hybrid organic-inorganic metal complexes are a versatile
class of materials, well-suited to structural design, by
modulation of both the ionic framework and the organic
constituent. Metal organic-phosphonate/phosphinates repre-
sent a very important series, as they have found application
in various fields (catalysis, ion exchange, material science,
(1) Vioux, A.; Le Bideau, J.; Mutin, P. H.; Leclercq, D. Top. Curr. Chem.
2004, 232, 145-174 and references therein.
(2) Allcock, H. R. AdV. Mater. 1994, 6, 106-115.
(3) Thompson, M. E. Chem. Mater. 1994, 6, 1168-1175.
(4) Clearfield, A. Curr. Opin. Solid State Mater. Sci. 2002, 6, 268-278.
(5) Clearfield, A. Curr. Opin. Solid State Mater. Sci. 2002, 6, 495-506.
(6) Bakhmutova, E. V.; Ouyang, X.; Medvedev, D.G.; Clearfield, A. Inorg.
Chem. 2003, 42, 7046-7051.
* To whom correspondence should be addressed. E-mail:
annabella.bianchi@iccom.cnr.it.
† Centre for Chiral and Molecular Technologies, Deakin University.
‡ Present address: Institut fu¨r Chemie, Freie Universita¨t Berlin, Fabeck-
strasse 34-36, 14195 Berlin, Germany.
(7) Stock, N.; Frey, S. A.; Stucky, G. D.; Cheetham, A. K. J. Chem. Soc.,
Dalton Trans. 2000, 4292-4296.
§ Dipartimento di Chimica, Universita` di Perugia.
| Istituto di Chimica dei Composti Organometallici, CNR.
Dipartimento di Chimica, Universita` di Firenze.
(8) Burlholder, E.; Golub, V. O.; O’Connor, C. J.; Zubieta, J. Inorg. Chim.
Acta 2002, 340, 127-132.
9416 Inorganic Chemistry, Vol. 44, No. 25, 2005
10.1021/ic050683p CCC: $30.25
© 2005 American Chemical Society
Published on Web 11/02/2005

Hybrids of the p,p′-Diphenylmethylenediphosphinate, pcp2-
Chart 1
compounds seems to be determinant not only for the
dimensionality but also for the rate of crystal growth. These
solids are, in fact, microcrystalline, and it was only possible
to characterize their structures thanks to the application of
the “ab initio” X-ray powder diffraction (XRPD) methods,
which represent a powerful analytical tool when single
crystals are not available. A 119Sn MAS NMR (magic-angle
spinning NMR) study has been performed on the tin
complex, whereas magnetic measurements were carried out
on the copper derivative.
the perception of there being less potentiality for building
up extended structures. However, recently we have exploited
the ligand capacity of the bifunctional p,p′diphenylmethyl-
enediphosphinate ligand, pcp2- (Chart 1),11 which, as far as
we know, has only previously been used once, to prepare a
metal complex, namely polymeric Ti(pcp)2, which was of
unspecified structure.12 This pcp2- dianion contains four
oxygen donor atoms, potentially allowing the coordination
of a single metal ion (either with one or two oxygens) and/
or bridging of multiple metal centers. Only the chelation of
a single metal ion by the two oxygens at the same PO2 seems
to be excluded because of the large bite angle. Therefore,
the formation of pcp-metal complexes displaying a remark-
able array of different structural arrangements could be
expected.
Experimental Section
Materials and Methods. All reagents were analytical-grade
commercial products and were used without further purification.
The p,p′-diphenylmethylenediphosphinic acid (H2pcp) was prepared
as previously described.11,15 The hydrothermal reactions were
performed in a stainless steel autoclave, with a Teflon insert (ca.
20 mL capacity), constructed in-house.
Synthesis of [Sn(pcp)], 1. A suspension of SnCl2‚2H2O (500
mg, 2.22 mmol) and CH2(PhP(O)(OH))2 (H2pcp) (656 mg, 2.22
mmol) was refluxed for 12 h in water. The solution was filtered
while still hot, before approximately one-third of the water was
removed in a vacuum. Upon cooling, a colorless precipitate formed,
which was filtered, washed with water, and dried in a vacuum.
Yield 810 mg, 88%. mp > 300 °C. Anal. Calcd for C13H12O4P2Sn:
C, 37.81; H, 2.93. Found: C, 37.20; H, 2.76%.
Synthesis of [Cu(pcp)], 2. The hydrothermal reaction of
Cu(O2CMe)2‚H2O (27 mg, 0.135 mmol) and H2pcp (40 mg, 0.135
mmol) in 6 mL of H2O, at 453 K for 3 days, followed by slow
cooling at room temperature, produced thin pale-blue crystals of
the complex. The compound was filtered, washed with water, and
dried in air, at room temperature.The complex is insoluble in water,
where it remains unchanged (any of the two hydrated [Cu(pcp)-
(H2O)2]‚H2O and [Cu(pcp)(H2O)2] complexes, previously synthe-
sized by conventional reactions, cannot be obtained starting from
the title complex).15 Yield 31 mg, 64%. Anal. Calcd for C13H12-
CuO4P2: C, 43.65; H, 3.38. Found: C, 43.58; H, 3.45.
X-ray Powder Diffraction Analysis. An X-ray powder diffrac-
tion pattern of [Sn(pcp)] was collected using a Bruker D8 advance
powder diffractometer, equipped with Cu KR radiation and operat-
ing in θ-2θ Bragg Brentano geometry at 40 kV and 30 mA. The
“SolX” solid-state detector was used. 0.8 mm divergence, 0.2
antiscatter, and 0.1 mm receiving slits were used. C/Ni Goebel
mirrors for the incident beam were used. The X-ray powder
diffraction pattern of [Cu(pcp)] was collected using a Philips
XPERT APD PW 3020 goniometer, using the Bragg-Brentano
θ-2θ geometry and equipped with Cu KR radiation and a bent
graphite monochromator on the diffracted beam. The LFF tube
operated at 40 kV, 30 mA. To minimize preferred orientations, the
sample was carefully side-loaded onto an aluminum sample holder
with an oriented quartz monocrystal underneath. The diffraction
patterns for [Sn(pcp)] and [Cu(pcp)] were fitted using a Pearson
VII profile function, and the position of the first 20 lines (KR1
maxima) was used for the indexing procedures. Indexing procedures
were performed using the TREOR program18 giving a monoclinic
cell [M(20) ) 18, F(20) ) 49] for 1 and a triclinic cell [M(20) )
13, F(20) ) 32]19 for 2 as the best solutions. Their cell parameters
are shown in Table 1. Systematical absences of the class 0k0, k )
As a matter of fact, we have successfully prepared and
characterized hybrid polymers with different metal ions(II),
like Be(II),13 Mn(II),14 Co(II),14 Ni(II),14 Cu(II),15 Zn(II),
16
and Pb(II).17 In particular, the anhydrous [Zn(pcp)] and
[Pb(pcp)] complexes present polymeric arrangements, char-
acterized, respectively, by 2D layered and 1D columnar
structures. All the other complexes so far isolated contain
water solvent molecules. Nickel, cobalt, and manganese give
rise to the isomorphous series [M(pcp)(H2O)3]‚H2O, where
the structure is arranged in the form of 2D hydrogen-bonded
layers.14 An analogous structural network has been found
for the [Cu(pcp)(H2O)2]‚H2O complex. It is worth noting the
role played by the crystallization and coordination water
molecules in these hydrated materials, which controls the
extended architecture via the networks of hydrogen-bonding
interactions.15 The cementing power of the hydrogen bonding
is also exemplified in the extended [Be(pcp)(H2O)2] deriva-
tive, where each pcp2- anion chelates one metal center.13
Now we report the synthesis and characterization of two
new anhydrous pcp polymers containing the tin(II) and
copper(II) ions, namely [Sn(pcp)] and [Cu(pcp)].The absence
of crystallization and coordination water in these two
(9) Grohol, D.; Ginge, F.; Clearfield, A. Inorg. Chem. 1999, 38, 751-
756.
(10) Oliver, K. W.; Rettig, S. J.; Thomson, R. C.; Trotter, J.; Xuia, S. Inorg.
Chem. 1997, 36, 2465-2468.
(11) Garst, M. E. Synth. Commun. 1979, 9, 261-266.
(12) Dahl, G. H.; Block, P. B. Inorg. Chem. 1966, 5, 1394-1396.
(13) Cecconi, F.; Dominguez, S.; Masciocchi, N.; Midollini, S,; Sironi,
A.; Vacca, A. Inorg. Chem. 2003, 42, 2350-2356.
(14) Berti, E.; Cecconi, F.; Ghilardi, C. A.; Midollini, S.; Orlandini, A.;
Pitzalis, E. Inorg. Chem. Commun. 2002, 5, 1041-1043.
(15) Ciattini, S.; Costantino, F.; Lorenzo-Luis, P.; Midollini, S.; Orlandini,
A.; Vacca, A. Inorg. Chem. 2005, 44, 4008-4016.
(16) Cecconi, F.; Dakternieks, D.; Duthie, A.; Ghilardi, G. A.; Gili, P.;
Lorenzo-Luis, P. A.; Midollini, S.; Orlandini, A. J. Solid State Chem.
2004, 177, 786-792.
(18) de Wolff, P. M. J. Appl. Crystallogr. 1968, 1, 108-113.
(19) Werner, P. E.; Eriksson, L.; Westdhal, M. J. Appl. Crystallogr. 1985,
18, 367-370.
(17) Cecconi, F.; Ghilardi, C. A.; Midollini, S.; Orlandini, A. Inorg. Chem.
Commun. 2003, 6, 546-548.
Inorganic Chemistry, Vol. 44, No. 25, 2005 9417

Beckmann et al.
Table 1. Crystal Data and Refinement Details for 1 and 2
1
2
empirical formula
fw
cryst syst
space group
a/Å
b/Å
c/ Å
C13H12O4P2Sn
412.87
monoclinic
P21/c
11.2851(1)
15.4495(6)
8.6830(1)
107.546 (1)
C13H12O4P2Cu
357.63
triclinic
P1h
10.7126(4)
13.0719(4)
4.9272(3)
92.067(5)
95.902(7)
87.847(4)
685.47(4)
2
R/deg
â/deg
γ/deg
vol/Å3
1443.44 (9)
4
Z
T/°C
25
25
calcd density/g‚cm-3
pattern range, 2θ/deg
step scan increment, 2θ/deg
step scan time/s
no. of data points
no. of reflns
no. of params
no. of restraints
1.84
7.5-100
0.01
1.67
3-130
0.02
30
6296
2600
90
70
0.053
0.037
0.054
15
10547
4162
71
56
a
Rp
Rwp
RF
0.075
0.096
0.084
1.56
b
2c
ød
5.07
Figure 1. Rietveld plots for 1 and 2 in the 3-70 2ϑ region. Calculated,
observed, and differences curves are shown.
b
2
2
a Rp ) ∑|Io - Ic|/∑Io. Rwp ) [∑w(Io - Ic)2/∑wIo ]1/2
.
c RF2 ) ∑|Fo
-
Fc |/∑|Fo |. d ø ) [∑w(Io - Ic)2/(No - Nvar)]1/2
2
2
120(1)°). The statistical weight of these restraints was maintained
to a relatively high value as the refinement proceeded, and it was
not possible to set it to zero because of some unrealistic light atom
bond distances. Thermal factors were refined only for metals and
P atoms, while those of light atoms were set at a fixed value and
were not refined.
2n + 1, and h0l, l ) 2n + 1, suggested P21/c as the best space
group for 1. Structural determination for 1 and 2 was performed
using the FOX program.20 The program optimizes a structural model
described by the use of building blocks defined in terms of their
internal coordinates, such as bond lengths, bond angles, and dihedral
angles. Optimization was performed by comparing the XRPD
patterns calculated from randomly generated configurations and
using the “integrated Rwp” (iRwp) as the cost function. Trial structures
were generated using the “Parallel Tempering” algorithm.21 The
starting model was composed by the free metal atom and the pcp2-
diphosphinic group. Hydrogen atoms were omitted. Starting values
for bond lengths and angles were taken from similar systems found
in the literature and constrained within a standard deviation of 0.15
Å and 10°, respectively. Internal bond lengths and angles were not
optimized. All other dihedral angles were left to change freely.
Antibump restraints (distance M-P ) 2.6(5) Å) between the metal
and the phosphorus atoms were fixed in order to speed up the
optimization process. Rietveld refinement of the structures was
performed using the GSAS program (see Figure 1).22 First of all,
zero shift, cell parameters, background, and profile shape were
refined. A pseudo voigt profile function23 (six terms) with two terms
for the correction of asymmetry at the low-angle region was used.24
Then atomic coordinates and isotropic thermal factors were refined.
The following soft restraints between all the atoms were imposed:
Sn-O ) 2.05(5), P-O ) 1.56(5), and P-C ) 1.83(5) Å for 1
and Cu-O ) 1.95(5), P-O ) 1.56(5), and P-C ) 1.83(5) Å for
2, and the “ideal planar ring” restraint for the phenyl groups was
used (C-C distance ) 1.39(5) Å and aromatic bond angles )
MAS NMR Measurements. The MAS NMR spectra were
measured using a Joel Eclipse Plus 400 spectrometer (at 161.84
(31P) and 149.05(119Sn)) equipped with a 6 mm MAS probe.
Crystalline NH4H2PO4 (δ 0.95) and c-Hex4Sn (δ -97.35) were used
as secondary references against aqueous H3PO4 (90%) and SnMe4.
The 119Sn MAS NMR spectra were obtained using cross-polariza-
tion (contact time 5 ms and recycle delay 10 s). A tensor analysis
was performed on the 119Sn MAS NMR data using the computer
program DM Fit 200225 and was based on relative intensities of
the spinning sidebands. Definitions are δiso (ppm) ) -σiso
)
-(σ11+σ22+σ33)/3; ú (ppm) ) σ33 - σiso and η ) (σ22 - σ11)/(σ33
- σiso), where σ11, σ22, and σ33 are the principal tensor components
of the shielding anisotropy (SA), sorted as follows |σ33 - σiso| >
|σ11 - σiso| > |σ22 - σiso|. 31P MAS NMR δ: 19.5, 18.3. 119Sn
MAS NMR δiso: -685.0; ú: 484; η: 0.40; σ11: 346; σ22: 540;
σ33: 1169.
Magnetic Measurements. Magnetic susceptibility of a sample
powder of [Cu(pcp)] was measured between 2 and 300 K with
applied magnetic fields of 1 T using a Cryogenic S600 SQUID
magnetometer. Data were corrected for the magnetism of the sample
holder, which was determined separately in the same temperature
range and field, and the underlying diamagnetism of each sample
was estimated from Pascal’s constants. Magnetization meas-
urements were performed on the same sample at 1.8 K with a
field up to 6.5 T. An electron paramagnetic resonance (EPR)
spectrum was recorded with an X-Band Bruker Elexsys E500
spectrometer.
(20) Favre-Nicolin, V.; Cerny, R. J. Appl. Crystallogr. 2002, 35, 734-
743.
(21) Falcioni, M.; Deem, M. W. J. Chem. Phys. 1999, 110, 1754-
1766.
(22) Larson, A. C.; von Dreele, R. B. Generalized Crystal Structure Analysis
System; Los Alamos National Laboratory: Los Alamos, NM, 2001.
(23) Thompson, P.; Cox, D. E.; Hastings, J. B. J. Appl. Crystallogr. 1987,
20, 79-83.
(24) Finger, L. W.; Cox, D. E.; Jephcoat, A. P. J. Appl. Crystallogr. 1994,
27, 892-900.
(25) Massiot, D.; Fayon, F.; Capron, M.; King, I.; Le Calve`, S.; Alonso,
B.; Durand, J.-O.; Bujoli, B.; Gan, Z.; Hoatson, G. Magn. Reson.
Chem. 2002, 40, 70-76.
9418 Inorganic Chemistry, Vol. 44, No. 25, 2005

Hybrids of the p,p′-Diphenylmethylenediphosphinate, pcp2-
Figure 2. Perspective view of the asymmetric units of [Sn(pcp)] 1 (left) and [Cu(pcp)] 2 (right) polymers. ORTEP drawing with 30% probability ellipsoids.
Figure 3. Propagation of the polymer of [Sn(pcp)] along the z direction.
Phenyl rings omitted for sake of clarity.
Results and Discussion
Descriptions of the Structures. The X-ray study of
[Sn(pcp)] and [Cu(pcp)] revealed polymeric structures for
both the compounds. Their asymmetric units are similar
(Figure 2), with the only significant difference being the
relative orientation of the phenyl rings, which are in the anti-
Figure 4. Portion of the polymer of [Sn(pcp)] showing the likely
interactions interesting the lone pair of tin(II), viewed normal to [100].
and syn-positions in 1 and 2, respectively.
(E ) electron pair) has been described in other tin(II)
polymeric phosphates and phosphonates complexes: see, for
examples, [Sn(O3PCH2PO3]2-,26 Sn4(O3PCH2CH2CO2)2-
Despite having similar building units, the unique nature
of the two metal ions results in different processes of building
up of the extended networks. In particular in 1, each tin center
is surrounded by four oxygens atoms from three pcp2-
ligands, with one acting as bidentate and the other two acting
as monodentate, with each pcp2- ligand bridging three metals
ions. This results in a one-dimensional ribbon, where six-
membered Sn-O-P-C-P-O rings alternate with eight-
membered Sn-O-P-O-Sn-O-P-O rings, forming an
undulated ladder-like structure. The open space between the
ribbons is reduced by the presence of the organic part of the
ligand, namely, the phenyl rings, and additionally by the
tin(II) ion lone pair, whose presence is well-inferred from
the hemilectic coordination around the metal (Figures 3 and
4). As a matter of fact, if we assume the O21-Sn-O41 as
one trans basal angle, the tin(II) coordination sphere re-
sembles a distorted square pyramid, missing the fifth donor
in the basal plane. The lone pair should be likely positioned
opposite to the O3 donor, which actually displays the shortest
Sn-O bond (2.074(7) Å) (see Table 2). The Sn-O bond
lengths, ranging from 2.074 to 2.162 Å, are similar to
analogous distances reported in the literature.26-30 The
truncated square pyramidal Sn(II) coordination sphere SnO4E
28
(C2O4),27 and Sn2(PO4)(C2O4)0.5.
In considering the activity of the tin(II) lone pair, it seems
to be likely pointing toward the C12-C13 bond of the P1
phenyl ring, with distances Sn‚‚‚C of 3.70 and 3.72 Å,
respectively, suggesting that there is some interaction (sum
of the van der Waals ) 3.87 Å). This interaction does not
contribute to extending the structure dimensionality, because
it involves the lone pair and one phenyl ring of the same
ribbon (Figure 4). Anyway, if we consider the inter-ribbon
Sn‚‚‚Sn separation of 4.57 Å (only slightly larger than the
sum of the van der Waals radii (4.34 Å), a lone pair pointing
toward the tin of the adjacent ribbon could not be excluded.
In this case, the coordination sphere including the lone pair
would be better described as trigonal bipyramidal rather than
square pyramidal. In a recent paper by some of us, interac-
tions between Te and the C(sp2)-C(sp2) bonds of two
individual phenyl rings have been observed, which are due
(27) Stock, N.; Stucky, G. D.; Cheetham, A. K. Chem. Commun. 2000,
2277-2278.
(28) Natarajan, S. J. Solid State Chem. 1998, 139, 200-203.
(29) Zapf, P. J.; Rose, D. J.; Haushalter, R. C.; Zubieta, J. J. Solid State
Chem. 1996, 125, 182-185.
(26) Adair, B. A.; Neeraj, S.; Cheetham, A. K. Chem. Mater. 2003, 15,
1518-1529.
(30) Zapf, P. J.; Rose, D. J.; Haushalter, R. C.; Zubieta, J. J. Solid State
Chem. 1997, 132, 438-442.
Inorganic Chemistry, Vol. 44, No. 25, 2005 9419

Beckmann et al.
Table 2. Selected Bond Distances (Å) and Angles (deg) for 1a
bond
distance
bond
angle
Sn1-O1
Sn1-O2#1
Sn1-O3
Sn1-O4#2
P1-O1
2.118(6)
2.127(6)
2.074(7)
2.162(6)
1.482(8)
1.494(8)
1.783(7)
1.759(7)
1.550(8)
1.457(7)
1.762(7)
1.767(6)
O1-Sn1-O2#1
O1-Sn1-O3
O1-Sn1-O4#2
O2#1-Sn1-O3
O2#1-Sn1-O4#2
O3-Sn1-O4#2
O1-P1-O2
O1-P1-C1
98.3(6)
85.6(5)
72.0(3)
73.3(3)
164.1(5)
93.0(5)
P1-O2
P1-C1
114.8(7)
109.2(5)
110.1(6)
106.9(6)
110.2(6)
105.2(9)
111.7(6)
108.8(6)
102.9(5)
113.3(6)
115.8(6)
103.5(4)
141.6(6)
155.7(6)
137.5(7)
172.4(9)
118.2(6)
Figure 5. Propagation of the polymer of [Cu(pcp)] along the z direction.
Phenyl rings omitted for sake of clarity.
P1-C11
P2-O3
O1-P1-C11
O2-P1-C1
P2-O4
P2-C1
O2-P1-C11
C1-P1-C11
O3-P2-O4
O3-P2-C1
O3-P2-C21
O4-P2-C1
O4-P2-C21
C1-P2-C21
Sn1-O1-P1
Sn1-O2-P1
Sn1-O3-P2
Sn1-O4-P2
P1-C1-P2
P2-C21
a Symmetry transformations used to generate equivalent atoms: #1, x,
3
1
3
-y + /2, z - /2; #2, x, -y + /2, z +1/2.
Table 3. Selected Bond Distances (Å) and Angles (deg) for 2a
bond
distance
bond
angle
Cu1-O1
Cu1-O2#1
Cu1-O3
Cu1-O3#2
Cu1-O4#3
P1-O1
2.019(13)
1.885(12)
2.411(12)
1.942(11)
1.965(11)
1.438(10)
1.506(11)
1.843(10)
1.810(10)
1.444(10)
1.518(11)
1.842(11)
1.763(10)
O1-Cu1-O2#1
O1-Cu1-O3
O1-Cu1-O3#2
O1-Cu1-O4#3
O2#1-Cu1-O3
O2#1-Cu1-O3#2
O2#1-Cu1-O4#3
O3-Cu1-O3#2
O3-Cu1-O4#3
O3#2-Cu1-O4#3
O1-P1-O2
89.5(6)
95.3(7)
173.1(7)
85.7(6)
93.3(6)
95.9(6)
175.0(4)
89.4(6)
89.5(6)
Figure 6. Portion of the polymer of [Cu(pcp)], showing the packing of
the strands, with phenyl groups included (best view).
P1-O2
P1-C1
P1-C11
P2-O3
ligand. The intersections of the ribbons are constituted by
dimeric units of copper(II) ions, which display a Cu‚‚‚Cu
separation of ca. 3.1 Å. Each copper center is five-
coordinated, being surrounded by five oxygens from four
different pcp2- ligands to give a square pyramidal geometry.
In turn, each phosphinate links four copper atoms, using all
of its four oxygen donors in monodentate or bidentate
fashion. The propagation of the polymer along the z direction
is given in Figures 5 and 6. The complete packing diagram
of [Cu(pcp)] with view normal to [001] is reported in Figure
S2 (in Supporting Information), revealing the stacked ar-
rangement of phenyl groups, which project into the voids
between the polymeric strands.
A structural framework analogous to the copper struc-
ture has been recently reported for the lead(II) polymer
[Pb(pcp)],16 which shows the same polymeric array and
packing but some differences in the first coordination sphere
of the metal, because of the presence of an operative electron
lone pair on the lead atom. Bond distances and angles within
the coordination sphere of 2 (see Table 3) are instead in
agreement with those found in the hydrated [Cu(pcp)(H2O)2]‚
H2O and [Cu(pcp)(bipy)(H2O)] compounds, where the pres-
ence of water molecules extends the dimensionality through
hydrogen-bonding interactions.15 The major difference ob-
served in the latter complexes with respect to the dehydrated
copper compound is the shorter length of the Cu-O apical
bond (Cu-Oap 2.215(6) and 2.174(4) versus 2.411(12) Å,
88.6(6)
P2-O4
121.1(10)
109.4(9)
107.9(9)
107.8(9)
108.0(8)
100.8(6)
115.1(10)
111.4(11)
114.2(8)
103.5(7)
106.5(7)
105.1(6)
132.0(11)
131.9(9)
144.5(11)
140.9(9)
116.2(7)
P2-C1
O1-P1-C1
P2-C21
O1-P1-C11
O2-P1-C1
O2-P1-C11
C1-P1-C11
O3-P2-O4
O3-P2-C1
O3-P2-C21
O4-P2-C1
O4-P2-C21
C1-P2-C21
Cu1-O1-P1
Cu1-O2-P1
Cu1-O3-P2
Cu1-O4-P2
P1-C1-P2
a Symmetry transformation used to generate equivalent atoms: #1, x, y,
z -1; #2, -x + 2, -y + 1, -z; #3, -x + 2, -y + 1, -z + 1.
to the lone pair on Te.31 Finally, additional contacts in the
range 3.7-4.0 Å are present between P1- and P2-phenyl
groups belonging to adjacent polymeric units. The complete
packing diagram with view normal to [001] is reported in
Figure S1 (in Supporting Information).
The copper derivative [Cu(pcp)] 2 has a polymeric
monodimensional structure, formed by two intersecting
undulated ribbons of Cu atoms bridged by the phosphinate
(31) Beckmann, J.; Dakternieks, D.; Duthie, A.; Mitchell, C.; Schu¨rmann,
M. Aust. J. Chem. 2005, 58, 119-127.
9420 Inorganic Chemistry, Vol. 44, No. 25, 2005

Hybrids of the p,p′-Diphenylmethylenediphosphinate, pcp2-
119Sn and 31P MAS NMR Spectra of [Sn(pcp)]. The 119Sn
MAS NMR spectrum of [Sn(pcp)] shows an isotropic
chemical shift at δiso ) -685.0, which is substantially
different from that of SnO (-208).32 In both compounds,
the primary coordination sphere of the Sn atoms is defined
by four oxygen atoms (CN ) 4). The chemical shift
difference is consistent with there being significantly more
electron density around the tin atom in [Sn(pcp)]. Relatively
large anisotropy (ú) is observed for both [Sn(pcp)] (484) and
SnO (654) due to asymmetry in the valence electron cloud
about tin.32 The asymmetry (η) is larger for [Sn(pcp)] (0.40)
than for SnO (0.10), with the latter having a 4-fold axis
through the tin atom.32 The 31P MAS NMR shows two signals
of equal intensity at 19.5 and 18.3 ppm, indicating the
presence of two inequivalent phosphorus atoms, which is
consistent with the structure determined by XRPD. These
resonances are more deshielded than those of solid H2pcp
(32.8 pp) and the lead analogue, [Pb(pcp)] (24.5 and 31.4
ppm).16
Chart 2
respectively). The fact is not surprising, if we consider that
the oxygen positioned in apical position in the title compound
is shared between the two copper atoms of a dimeric subunit.
Finally, the fact that the isostructural copper and lead
derivatives display completely different structures compared
to the related tin polymer could be explained considering
that the 5s2 lone pair of tin(II) is stereochemically more active
than the 6s2 lone pair of lead(II) (the stereochemical activity
of the lone pair is more important for small cations than for
big cations: it increases going up a group in the periodic
table, in agreement with the evolution of the covalent
character of the elements). In this context, the stereochemical
requirement of the lone pair on tin(II) seems be important
in the construction of the polymer, preventing a more
compact structure. Finally, the [Zn(pcp)] hybrid material
recently reported16 displays a 2D layered structure, com-
pletely different from the networks found in the tin, lead,
and copper derivatives.
Magnetic Properties of [Cu(pcp)]. From a magnetic point
of view, [Cu(pcp)] can be regarded as a spin ladder
compound, with the rail being the Cu-O-P-O-Cu skel-
eton and the legs being the O bridge between centro-
symmetrically related copper ions. A constant value of
øT ) 0.44 emu K mol-1 is observed for 2 down to 50 K, as
expected for a simple S ) 1/2 ion with gave ≈ 2.19. This is
in agreement with previously reported g values on systems
containing similar cromophores33 and has been confirmed
by room temperature X-Band EPR. Indeed, the spectrum
clearly shows the axial signal expected for a square pyramidal
2
Cu(II) center, with g| ) 2.40 and g ) 2.09 (gave) x[(2g
+ g|2)/3] ) 2.19), indicating the unpaired electron resides
2
2
in a dx -y orbital, as expected on the basis of the structural
features.34 Below 50 K, a decrease in øT is clearly visible,
indicating the presence of weak antiferromagnetic interac-
tions (Figure 8). The magnetic data at low temperature were
fitted by using two different models. In the first one, we
assumed a simple dimer interaction (H ) JS1S2) with
molecular field correction to account for the interaction along
the rail of the ladder:35
Topological Analysis of the 1 and 2 Networks.We have
wondered whether 1 and 2 complexes are topologically
equivalent or not, that is, whether an interconversion between
their skeletons is possible. With this in mind, we have
potentially reduced the two metal centers at the same
coordination number, breaking up one Cu-O bond, namely,
(either one of the two) Cu-O3, to generate the two copper
models, as shown in Chart 2.
Among these, we restricted our attention to model 2b,
which, being constructed by a copper linked to three different
pcp’s, seems more similar than 2a to the tin model. Going
ahead with the schematization, the carbon and oxygen nodes
in both of the models (tin and copper 2b) were suppressed,
to obtain two new skeletons (1′ and 2′), respectively
characterized by three- and four-membered and three-and
six-membered rings (see Figure 7). At this point, the different
topology of the two models is just pointed out. However, a
further condensation could be performed, by suppressing the
three-membered rings. Thus, two models of similar topology
were obtained, where the ligand is reduced at a 3-connected
point. The new skeletons, 1′′ and 2′′, again evidence the
difference between the two models, which stems from the
different sequence by which the ligand disposes itself with
respect to the metal centers (in 1′′, the chelating bite connects
the two parallel lines, while in 2′′, it lies on the lines
themselves).
øM
ø′M )
(1)
(1 + (zJmf/g2â2)øM)
Here, Jmf is the interdimer interactions, z ) 2 is the number
of nearest neighbors, and øM is the dimer susceptibility value
calculated as
Ng2â2
kT(3 + exp(J/kT))
øM )
(2)
This yielded a fit with best parameter values g ) 2.195 (
(32) Cossement, C.; Darville, J.; Gilles, J. M.; Nagy, J. B.; Fernandez, C.;
Amoureux, J. P. Magn. Reson. Chem. 1992, 30, 263-270.
(33) Cini, R.; Colamarino, P.; Orioli, P. L.; Smith, L. S.; Newman, P. R.;
Nannelli, P.; Gillmann, H. D. Inorg. Chem. 1977, 16, 3223-3226.
(34) Bencini, A.; Gatteschi, D. In Inorganic electronic structure and
spectroscopy; Lever, A. B. P., Solomon, E. I., Eds.; Wiley-Interscience:
New York, 1999.
(35) O’Connor, C. J. Prog. Inorg. Chem. 1982, 29, 203-283.
Inorganic Chemistry, Vol. 44, No. 25, 2005 9421

Beckmann et al.
Figure 7. Transformation of the Sn and Cu networks 1 and 2b into an equivalent 3-connected net. In the first step, the carbon and oxygen nodes (white
balls in 1 and 2b) are suppressed to give two topologically different networks 1′ and 2′. A further reduction of the pcp ligand into a 3-connected point gives
two nets 1′′ and 2′′, in which, respectively, the chelating bite of the ligand (represented in bold) connects the two parallel lines or lies on the lines themselves.
0.003, J ) 3.17 ( 0.04 cm-1, and Jmf ) 0.57 ( 0.02 cm-1
(R2 ) Σi(øobsd(i)Ti - øcalcd(i)T)2/øcalcd(i)Ti2 ) 7.6 × 10-6).
Even if J/Jmf < 10 so that the molecular field approximation
cannot be considered completely valid, the obtained J value
for the intradimer coupling is in good agreement with the
results of the M vs H curve measured at 1.8 K (Figure 9),
which shows a singlet-triplet field induced transition around
4 T, suggesting J ≈ 4 cm-1.
A more refined model for the fit involved the application
of a fourth order polynomial formula, developed for calculat-
ing the susceptibility of spin ladders:36,37
øM ) (Ng2â2/4kT)
Figure 8. Plot of øT vs T for [Cu(pcp)] and best fit curves obtained on
2
application of the two models (the two best fit curves are indistinguishable).
In the inset, the ø vs T plot at low T is reported together with the curves
calculated from best fit parameters obtained by fitting øT experimental
points.
1
kT
1
4
1
2
1
kT
1
16
1
4
1 +
- J′ - J′′ +
-
J′2 + J′J′′ +
(
)
(
) (
)
3
1
kT
1
192
1
1
1
1
+
J′3 + J′2J′′ - J′2J′′ - J′3
+
(
) (
)
32
1
64
32
1
48
24
4
1
5
5
[
]
+
J′4 - J′3J′′ + J′2J′′2 - J′3J′′ +
J′3
(
) (
)
kT 768
48
384
(3)
Here, J′ is the leg interaction and J′′ is the rail one; best fit
curves were obtained with the following parameters: g )
2.193 ( 0.002, J′) 4.76 ( 0.03 cm-1, and J′′) 0.51 ( 0.02
cm-1 (R2 ) 1.45 × 10-6). Both fits are of good quality and
give essentially the same best fit parameters values. Interest-
ingly, both models reproduce equally well the position and
the amplitude of the maximum observed at low temperature
in the ø vs T plot (Figure 8, inset). On the other hand, this
Figure 9. Plot of M vs H curve for [Cu(pcp)], measured at 1.8 K.
(36) Zhou, P.; Drumheller, J. E.; Rubenbacker, G. V.; Halvorson, K.;
Willett, R. D. J. Appl. Phys. 1991, 69, 5804-5806.
(37) Rushbrooke, G. S.; Baker, G. A.; Wood, P. J. In Phase transition
critical phenomena; Domb, C., Green, M. S., Eds.; Academic Press:
New York, 1974; Vol. 3, pp 245-356.
did not occur if we neglected one of the two exchange-
coupling paths, by assuming either a simple dimer model
based on eq 2 alone or an isolated chain one based on the
9422 Inorganic Chemistry, Vol. 44, No. 25, 2005

Hybrids of the p,p′-Diphenylmethylenediphosphinate, pcp2-
Bonner Fisher expression38 for isotropic chains. This clearly
indicates that both interactions are important in determining
the magnetic behavior of 2.
As for the intradimer interaction, the small value obtained
for the coupling constant is not surprising, despite the short
distance separating the two Cu(II) ions, because the magnetic
orbital of the two copper ions are unfavorably oriented to
give rise to a strong interaction. Indeed, for each copper pair,
the oxygen which occupies the basal position of one ion is
bound in the axial position to the other one. As the unpaired
electron resides in the dx -y orbital, the degree of delocal-
ization from one site to the other one is then expected to be
quite small. In this case, a comparison with ref 40 evidences
a stronger interaction for 2, which can be related both to the
much shorter distance between Cu ions (3.1 Å vs 3.6 Å)
and to the somewhat shorter Cu-Oapical elongation (2.41 Å
vs 2.505 Å).
The observed interaction along the phosphinate group
obtained in 2 compares well with what has been reported
for similar systems.39-41 The small value of the constant is
probably much the result of different effects counterbalancing
each other. Indeed, earlier studies suggested three parameters
to govern the sign of magnetic interaction in these kind of
systems, namely, the inter-ion distance, the planarity of the
chromophore, and the puckering of the ring.39 Analysis of
these structural parameters evidences that, while the pucker-
ing of the ring is somewhat more pronounced here and the
Cu-Cu distance is shorter in 2 than in the ethyl derivatives39
(4.93 Å vs 4.95 Å and 5.03 Å), the planarity of the chro-
mophore is much more pronounced for 2. We note here that
the main structural feature differentiating 2 from complexes
reported in refs 39 and 40 is the value of the dihedral angle
between the CuO4 planes and the O-P-O one, which is
much smaller in 2; we then suggest this might be another
relevant parameter influencing the degree of magnetic
coupling observed in copper-phosphinate complexes.
2
2
On the basis of this discussion, we can then conclude that
both couplings are weak, antiferromagnetic, and in reasonable
agreement with reported literature data.33,39-41
Acknowledgment. Thanks are expressed to Dr. Samuele
Ciattini (Universita` di Firenze) for the collection of the X-ray
powder spectra for[Sn(pcp)].
(38) Kahn, O. In Molecular magnetism; Wiley-VCH: New York, 1993.
(39) Haynes, J. S.; Oliver, K. W.; Rettig, S. J.; Thompson, R. C.; Trotter,
J. Can. J. Chem. 1984, 62, 891-898.
Supporting Information Available: X-ray crystallographic data
in CIF format; Figures S1 and S2, reporting the complete packing
diagrams of 1 and 2, in tiff format.
(40) Oliver, K. W.; Rettig, S. J.; Thompson, R. C.; Trotter, J.; Xia, S. Inorg.
Chem. 1997, 36, 2465-2468.
(41) Haynes, J. S.; Oliver, K. W.; Thompson, R. C. Can. J. Chem. 1985,
63, 1111-1117.
IC050683P
Inorganic Chemistry, Vol. 44, No. 25, 2005 9423