A growing family: New structures of coordination polymers containing adamantane-shaped phosphorus-nitrogen cage ligands

Article abstract of DOI:10.1016/j.ica.2010.08.052

Two new structurally characterized coordination polymers containing the P₄(NR)₆ ligand system are described. A convenient one-pot synthesis of P₄(NR)₆ (R = benzyl) via reaction of lithiated primary amine with ph

Full text of DOI:10.1016/j.ica.2010.08.052

Inorganica Chimica Acta 364 (2010) 261–265
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Note
A growing family: New structures of coordination polymers containing
adamantane-shaped phosphorus–nitrogen cage ligands
Alex B. Spore a, Natalie M. Rizzo a, Bruce C. Noll b,1, Roger D. Sommer a,
⇑
a Department of Chemistry, DePaul University, 1110 W. Belden Ave., Chicago, IL 60614, USA
b Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556-5670, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 27 October 2010
Two new structurally characterized coordination polymers containing the P4(NR)6 ligand system are
described. A convenient one-pot synthesis of P4(NR)6 (R = benzyl) via reaction of lithiated primary amine
with phosphorus trichloride demonstrates an expanded scope for the preparation of this adamantane-
type structure. Reactions of P4(NR)6 (R = Et, Bn) with cuprous iodide yield different products due to the
differences in steric demands of the ligands.
Dedicated to Arnold L. Rheingold
Keywords:
Ó 2010 Elsevier B.V. All rights reserved.
Coordination polymers
Phosphorus–nitrogen
Copper
1. Introduction
od shown in Scheme 2. This adaptation of synthetic routes used to
build some phosphazane macrocycles minimizes the amount of
Phosphorus(III)–nitrogen compounds (phosphazanes) are well
known as effective P-donor ligands in coordination chemistry and
host–guest chemistry. Recent work has shown that P–N com-
pounds also display interesting macrocyclic structures and form ex-
tended structures with metal ions to make their own types of
coordination polymers and networks [1–5]. A long-known reaction
of phosphorus trichloride with methylamine yields the adaman-
tane shaped cage molecule shown in Scheme 1. As simple as this
reaction seems, the family of P4(NR)6 derivatives that have been
isolated remains very small. Only two (methyl and isopropyl) deriv-
atives of this cage have been structurally characterized [6,7]. The
R = ethyl derivative has been observed by {1H}31P NMR, but not
structurally characterized [8]. The typical preparation shown in
Scheme 1 uses excess amine to absorb the hydrogen chloride
evolved during the reaction. This can result in a large amount of
amine hydrochloride, mixed with a small amount of the desired
product. As the alkyl chains are extended (n-propyl. . .) the hydro-
chloride salt by-products have significant solubility in organic sol-
vents. This complicates the workup considerably. Still worse, the
reaction no longer forms the desired cage exclusively, but a mixture
dominated by RN@P(–NHR)2 and P(–NHR)3 [9,10]. A more elaborate
preparation is reported for the isopropyl analogue [11].
amine required, and eliminates the presence of long-chain hydro-
chloride salts in the workup [12]. The first step of the process uses
lithiated amine both as the nucleophile to form P–N bonds, and the
base to neutralize liberated HCl. Stoichiometry prevents all P–Cl
bonds from reacting, and allows the formation of the precursors
(RNH-PCl2), (RNPCl)2 and (RNPCl)3. Later addition of triethylamine
liberates the remaining amine from the hydrochloride salt to com-
plete cyclization to P4(NR)6.
As part of the current effort to explore the structure and coordi-
nation chemistry of these ligands, X-ray crystallographic studies of
two new P4(NR)6/CuI coordination compounds are reported. Coor-
dination complexes prepared from the new P4(NR)6 ligands show
the tunability of this system with respect to steric crowding
around the P donor atoms. The core of this ligand can accommo-
date moderately bulky side chains that may be useful in dictating
the chemical and physical properties of complexes formed by this
cage ligand. The work below demonstrates one such case.
2. Experimental
General considerations: All reactions were carried out under
nitrogen atmosphere, either using Schlenk techniques, or working
in a nitrogen glovebox. Tetrahydrofuran, toluene and hexanes were
all distilled from Na/benzophenone before use. Anhydrous diethyl
ether and acetonitrile were used as purchased after storing over
molecular sieves. Anhydrous ethylamine was purchased from Sig-
ma–Aldrich and used as purchased. Higher amines were dried over
potassium hydroxide or calcium hydride and distilled before use.
Proton and phosphorus NMR spectra were acquired on a Bruker
This report details the synthesis of a new member of the
P4(NR)6 family according to a two-step, one-pot preparation meth-
⇑
Corresponding author. Tel.: +1 773 325 7322; fax: +1 773 325 7421.
E-mail address: rsommer@depaul.edu (R.D. Sommer).
Present address: Bruker-AXS Inc., 5465 E. Cheryl Parkway, Madison, WI 57311,
1
USA.
0020-1693/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2010.08.052

262
A.B. Spore et al. / Inorganica Chimica Acta 364 (2010) 261–265
Avance 300 NMR system (121.5 MHz for 31P) in d8 toluene, and
referenced to toluene or 85% H3PO4 external standard, respectively.
immediately upon addition, then disappeared with stirring. Color-
less crystals formed after 2 days. Suction filtration isolated 0.155 g
of product (57%). Anal. Calc. for: C, 25.16; H, 5.28; N, 14.68; Cu,
10.4; P, 21.6. Found: C, 25.84; H, 4.75; N, 14.63; Cu, 11.1; P, 20.6%.
2.1. Preparation of 1a, method 1
Phosphorus trichloride (15.0 mL, 0.175 mol) was added drop-
wise with mechanical stirring to 70.0 mL (1.06 mol) of ethylamine
at À60 °C. The yellow/white reaction mixture was allowed to warm
to room temperature. The reaction mixture was then heated to
150 °C to a give a yellow liquid. Cooling the mixture left an amber
oily liquid that formed a paste upon standing. Hexanes (3 Â 50 mL)
were added to extract the product. Filtration and removal of sol-
vent under vacuum yielded a colorless oil that solidifies on stand-
ing. (31P NMR, 79.5 ppm). Yield: 3.812 g (26.6%). Anal. Calc. for: C,
37.7.0; H, 7.91; N, 21.99. Found: C, 37.97; H, 7.48; N, 21.77%..
2.4. Preparation of 3, [P4(NBn)6(CuI)2(CH3CN)2ÁCH3CN]n
A solution of cuprous iodide (0.126 g, 0.662 mmol) in 6 mL of
acetonitrile was added to
a solution of ligand 1b (0.451 g,
0.598 mmol) in 10 mL of acetonitrile. X-ray quality crystals of
product grew after standing for 3 days. The supernatant was re-
moved by pipet, and the crystals washed with 3Â1 mL of acetoni-
trile. Yield: 0.130 g, 30% based on Cu. The crystals of 3 desolvate
rapidly, giving a white powder with composition consistent with
P4(NBn)6(CuI)2(CH3CN)2. Anal. Calc. for: C, 45.37; H, 3.97; N, 9.20;
Cu, 10.44; P, 10.17. Found: C, 46.12; H, 3.75; N, 8.55; Cu, 10.44;
P, 10.74%.
2.2. Preparation of 1b, method 2
A 250 mL roundbottom flask equipped with a stir bar and
rubber septum was charged with 40 mL of tetrahydrofuran,
and 1.88 mL (17.2 mmol) of benzylamine. After cooling the solu-
tion to 0 °C, 10.8 mL of 1.6 M n-butyllithium were added by syr-
inge. The resulting pink slurry was allowed to warm to room
temperature over 1 h. A second round bottom flask containing
20 mL of tetrahydrofuran and 1.0 mL (11.5 mmol) of phosphorus
trichloride was cooled to À60 °C in a dry ice/2-propanol bath,
and treated dropwise with the lithiated amine. The temperature
was maintained for 7 h before the reaction was allowed to warm
to 20 °C for an additional 12 h. A syringe was used to add
1.00 mL of triethylamine and the reaction stirred for an addi-
tional 24–36 h. A second portion of triethylamine was added
and the reaction stirred for another 72 h. Solvent was removed
under vacuum and the resulting white paste suspended in dry
toluene. Filtration and removal of solvent resulted in a yellow
oil. Column chromatography through silica gel, followed by re-
moval of solvent under vacuum yields a colorless oil (31P NMR,
80.9 ppm). Yield: 0.69 g, 32%. The ligand was typically used
immediately after preparation. Elemental analyses were low for
C and N. This was attributed to the possible oxidation during
handling. Anal. Calc. for: C, 66.84; H, 5.61; N, 11.14; Found: C,
63.51; H, 5.66; N, 10.06%.
2.5. X-ray Crystallography
Crystals were handled under oil and the specimen crystals were
affixed to the aperture of a MiTeGen MicroMount mounted to a ta-
pered copper pin. This assembly was transferred to the goniometer
of a Bruker CCD diffractometer equipped with an Oxford Cryosys-
tems low-temperature apparatus operating at 100 K (200 K for 3).
The unit cell and orientation matrix were determined using reflec-
tions harvested from three sets of 12 0.5° / scans, such that
I > 20r(I). Final unit cell parameters were determined using reflec-
tions harvested from each entire data set. All data were corrected
for Lorenz and polarization effects, as well as for absorption. Struc-
ture solution and refinement utilized the SHELXTL software package
as implemented in the APEX2 software suite [13]. Figures were con-
Table 1
Summary of crystal data for compounds 2 and 3.
Compound
2
3
CSD#
Formula
Fw
Space group
a (Å)
b (Å)
775 613
C24H60Cu2I2N12P8
775 614
C48H51Cu2I2N9P4
1258.74
P21/n
12.9771(12)
19.0563(16)
21.011(2)
90
92.626(3)
90
5190.5(8)
4
1145.48
P21/n
11.4755(4)
34.0911(12)
11.9048(4)
90
111.712(2)
90
4326.9(3)
4
2.3. Preparation of 2, [P4(NEt)6CuI]n
c (Å)
Cuprous iodide (0.091 g, 0.48 mmol) was dissolved in 10 mL of
dry acetonitrile and added dropwise to a solution of 0.353 g
(0.92 mmol) of 1a in 15 mL acetonitrile. A white precipitate formed
a
(°)
b (°)
c
(°)
V (Å3)
Z
qcalc (g/cmÀ3
F(0 0 0)
)
1.758
2288
2.741
1.611
2504
2.175
l
(Mo K
a
) (mmÀ1
)
k (Å)
Temperature (°C)
R; wR
0.71073
À173
0.71073
À73
0.0469; 0.1111
1.305
0.0406; 0.1164
1.044
Goodness-of-fit (GOF)
Scheme 1. A simple preparation of the P4(NR)6 cage (method 1) that only works
well for R = CH3 and C2H5.
Scheme 2. Reaction to prepare P4(NR)6 (R = benzyl) (method 2). {1H}31P NMR spectra of reaction mixture are included in supplementary material.

A.B. Spore et al. / Inorganica Chimica Acta 364 (2010) 261–265
263
structed and rendered using ORTEP-3 for Windows [14] and POV-RAY
3. Results and discussion
for Windows 3.6 [15]. A summary of crystallographic data for all
compounds is provided in Table 1 and details of the crystallogra-
phy are provided in Supplementary material.
3.1. New ligands
The preparation of P4(NBn)6 can be achieved by reaction of the
lithiated amine with phosphorus trichloride. When phosphorus tri-
chloride is treated with a suspension of lithiated primary amine in
a ratio of 2–3, the reaction initially forms a mixture of BnNHPCl2
(doublet d 158 ppm, J = 50 Hz) and trimeric (BnNPCl)3 (d 134, 125
and 104 ppm) and (BnNPCl)2 (d 227) [16,17]. Treatment of this
mixture with triethylamine deprotonates the hydrochloride
byproduct from the first step to complete the slow formation of
the P4(NR)6 structure. Monitoring of the reaction by 31P NMR
shows that the reaction progresses through a complicated series
of intermediates that collapse to the appropriate resonance near
80 ppm. Numerous unidentified intermediates are evident, with
resonances that are not consistent with previously suggested pre-
cursors to the P4(NR)6 cage. It is noteworthy that yields and purity
of products from this reaction are best when there is a slight excess
of PCl3. This is easily achieved by using slightly less than the re-
quired amount of butyllithium and triethylamine. It is not clear
from this study what route is taken to reach the P4(NR)6 structure.
The 31P NMR spectra of reaction mixtures show resonances appro-
priate for the coexistence of multiple intermediates, with no clear
distinction of whether dimer, trimer, or both are the direct precur-
sor of P4(NBn)6. This aspect of the experiment warrants its own de-
tailed study.
Table 2
Selected bond lengths and angles for compounds 2 and 3.
2
3
Cu1–P5
Cu1–P1
Cu1–I1
Cu2–P7
Cu2–P3(i)
Cu2–I2
2.2465(8)
2.2508(8)
2.5336(4)
2.2406(8)
2.2451(8)
2.5268(4)
Cu1–N7
Cu1–P1
Cu1–I1(i)
Cu1–I1
Cu2–N8
Cu2–P2
Cu2–I2(ii)
Cu2–I2
1.998(4)
2.2170(13)
2.6567(7)
2.6922(7)
2.010(5)
2.2117(13)
2.6346(8)
2.6857(7)
P5–Cu1–P1
P5–Cu1–I1
P1–Cu1–I1
P7–Cu2–P3
P7–Cu2–I2
P3–Cu2–I2
124.79(3)
117.65(3)
117.53(2)
127.17(3)
119.06(2)
113.72(2)
N7–Cu1–P1
N7–Cu1–I1(i)
P1–Cu1–I1(i)
N7–Cu1–I1
P1–Cu1–I1
I1–Cu1–I1(i)
N8–Cu2–P2
N8–Cu2–I2
P2–Cu2–I2
N8–Cu2–I2
P2–Cu2–I2
I2–Cu2–I2(ii)
Cu1–I1–Cu1(i)
Cu2–I2–Cu2(ii)
116.24(13)
104.08(12)
118.81(4)
102.75(13)
106.05(4)
107.52(2)
119.12(13)
106.12(14)
112.56(4)
102.77(13)
115.37(4)
98.43(2)
72.48(2)
81.57(2)
3.2. Preparation and structural characterization of CuI coordination
polymers
All crystals were grown from acetonitrile reaction mixtures.
Bond lengths and angles for the Cu coordination sphere, shown
in Table 2, and P4(NR)6 cage are consistent with other known struc-
tures. A slight shortening of P–N bond lengths is noted when P is
bound to copper. Some significant lengthening of P–N bond dis-
tances is noted in 3.
3.2.1. Structure of 2, [P4(NEt)6CuI]n
Compound 2 crystallizes in the monoclinic space group P21/n.
The asymmetric unit contains two cage ligands and two trigonal
planar copper ions as shown in Fig. 1. Both copper ions are coordi-
nated by two P donor atoms and a terminal iodide. The polymer
chains run parallel to the a-axis of the unit cell as shown in
Fig. 2. One ethyl group, attached to N1, shows evenly distributed
positional disorder over two sites. Calculated cone angles for
P4(NEt)6 were 157° for P atoms bound to Cu [18]. The P–N bonds
within the cage fall in a normal range for those in previously deter-
mined structures with an average for 2 of 1.708 Å. This suggests lit-
Fig. 1. Thermal ellipsoid plot of the asymmetric unit of 2. Ellipsoids drawn at 50%
probability. Hydrogen atoms are removed for clarity.
Fig. 2. Polymer segment of 2 along a-axis. Symmetry operations: (i) x, y, z À 1 (ii) x, y, z + 1. Thermal ellipsoids drawn at 50% probability. H atoms removed for clarity.

264
A.B. Spore et al. / Inorganica Chimica Acta 364 (2010) 261–265
of about 2:1. Adjacent to the disordered benzyl group is one unco-
ordinated acetonitrile solvent, also disordered over two orienta-
tions. The tetrahedral coordination environment of Cu is
completed by a bound acetonitrile as the fourth ligand. Steric
crowding near P prevents multiple cage ligands from binding Cu.
As a consequence, the 1:1 reaction of P4(NBn)6 and CuI produces
a linear coordination polymer that propagates by formation of a
Cu2I2 dimer as shown in Fig. 4. The Cu–Cu distances of 3.146 and
3.476 Å for Cu1 and Cu2, respectively, suggest no Cu–Cu bonding
is present. The Cu–P distances are comparable to other P4(NR)6
complexes. The P–N bond lengths and angles in the core of the li-
gand also display a typical distribution of values, with the excep-
tion of the P3–N3 and P4–N5 bonds. These bonds undergo
significant distortion, deviating from the average P–N bond dis-
tance by 0.047 and 0.026 Å, respectively. The N atoms of the cage
are generally planar, with the greatest deviations seen for N3 and
N5 where the sums of angles were 346° and 350°, respectively.
The presence of benzyl groups on the P4N6 core significantly in-
creases steric crowding. Calculated estimates of the cone angles for
P1 and P2 are 185° and 179°, respectively. Such crowding serves to
limit to one the number of ligands that may fit around a metal cen-
ter. Fig. 4 also shows how the benzyl groups near P1 and P2 are ori-
ented away to allow coordination of the copper atoms and form
the pocket that accommodates the acetonitrile ligand. While in-
tra-chain
p
–
p
interactions between benzyl groups and bound ace-
-stacking
tonitrile are observable, there is little evidence for
p
Fig. 3. Asymmetric unit of 3. Thermal ellipsoids of Cu, I, N and P drawn at 50%
probability. Benzyl group carbons drawn as stick models for clarity. Hydrogen
atoms removed for clarity.
between phenyl rings – either within or between chains. The close
proximity of the benzyl groups around the cage requires that if any
one benzyl group changes its conformation, at least one other must
move to accommodate it. Estimates of the steric crowding at P3
and P4 were calculated by placing dummy atoms at a distance of
2.24 Å from P and computing cone angles as for P–Cu bonds. The
estimated values for these cone angles were 215° and 211°, respec-
tively. The contrast in cone angles for structures 2 and 3 is shown
in Table 3. Such crowding coincides with the distortion in the core
structure of the ligand. The lengthening of the P3–N3 and P4–N5
bonds coincides with the closest approach of benzyl groups to P3
and P4 atoms.
tle change to the ligand core as a result of larger R groups. The N
atoms in the cage ligands are generally close to planar. The average
for the sum of angles around N is 354°, with a minimum value of
346°.
3.2.2. Structure of 3, [P4(NBn)6(CuI)2(CH3CN)2ÁCH3CN]n
Crystals of 3 grow in the monoclinic space group P21/n from a
solution of hot acetonitrile. Fig. 3 shows that the asymmetric unit
consists of one cage ligand bridging two metal centers. The benzyl
group attached to N3 occupies two sites with relative occupancies
Without P–Cu–P links, compound 3 is moderately soluble. It can
be recrystallized from hot acetonitrile with minimal decomposi-
Fig. 4. Polymer segment of 3. Symmetry operations: (i) Àx + 2, Ày + 2, Àz + 2 (ii) Àx + 2, Ày + 1, Àz + 2. Thermal ellipsoids drawn at 50% probability. Hydrogens removed and
benzyl groups shown as ball and stick for clarity.

A.B. Spore et al. / Inorganica Chimica Acta 364 (2010) 261–265
265
Table 3
(Bruker Avance 300) acquired under National Science Foundation
CCLI A&I program Grant No. DUE- 0310624. X-ray instrumentation
acquired under US NSF grant CHE-0443233 (University of Notre
Dame).
Summary of cone angles for coordinated and non-coordinated P atoms in 2 and 3.
Compound, R
Donor
atoms
Cone angle
P–Cu–H°
Cone angle
P-Dummy-H°
a
b
2, C2H5
3, CH2(C6H5)
2P, IÀ
157
182
160
213
P, 2IÀ, CH3CN
Appendix A. Supplementary material
a
Calculated from crystals structure data based on P–Cu–Hphenyl (on ligand) angle
(Ref. [18]).
b
CCDC 775613 and 775614 contain the supplementary crystallo-
graphic data for compounds 2 and 3 respectively. These data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/conts/
retrieving.html, or from the Cambridge Crystallographic Data Cen-
tre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-
033; or e-mail: deposit@ccdc.cam.ac.uk X-ray. NMR spectra from
reaction are also supplied as supplemental figures. Supplementary
data associated with this article can be found, in the online version,
at doi:10.1016/j.ica.2010.08.052.
Dummy atom placed at 2.24 Å and angle computed identical to P-donor atoms.
tion. Crystals grown in such a manner had identical structure to
those originally isolated from the reaction mixture. The steric
crowding of the non-coordinated P atoms may not only limit the
number of ligands that can bind to a cuprous ion, but also prevent
the P4(NBn)6 ligand from binding more than two metal ions.
4. Conclusion
Reactions of closo-tetraphosphorushexaalkylimide ligands
P4(NR)6 (R = ethyl, benzyl) with cuprous iodide has afforded two
new structures, expanding the group of structurally characterized
examples to four. The structural evidence in [P4(NEt)6CuI]n (2) sug-
gests that extending the length of the primary alkyl group does not
affect the structure of the P4N6 core. By modification of a well-
known reaction to form P–N clusters and cages, the preparation of
the sterically hindered P4(NBn)6 cage ligand can be achieved by a
simple process in moderate yield. The structural characterization
of the CuI complex 3 shows that this ligand can bind multiple metal
ions. But connectivity of 3 differs from 2 and [P4(NCH3)6CuI]n [4], in
part, due to congestion around the P4N6 core. Such observations sug-
gest a way to control the connectivity of coordination polymers and
networks that might be formed with this family of ligands.
References
[1] S.L. James, X. Xu, M. Nieuwenhuyzen, Polym. Prepr. 45 (2004) 423.
[2] S.L. James, X. Xu, R.V. Law, Macromol. Symp. 196 (2003) 187.
[3] X. Xu, M. Nieuwenhuyzen, S.L. James, Angew. Chem., Int. Ed. Eng. 41 (2002)
764.
[4] R.D. Sommer, A.L. Rheingold, Acta Crystallogr., Sect. E E62 (2006) m74.
[5] E.D. Leser, B.C. Noll, R.D. Sommer, Polyhedron 29 (2010) 2053.
[6] R.R. Holmes, J.A. Forstner, Inorg. Synth. 8 (1966) 63.
[7] O.J. Scherer, K. Andres, C. Krueger, Y.-H. Tsay, G. Wolmershaeuser, Angew.
Chem. 92 (1980) 563.
[8] T.G. Hill, R.C. Haltiwanger, M.L. Thompson, S.A. Katz, A.D. Norman, Inorg.
Chem. 33 (1994) 1770.
[9] O.I. Kolodiazhnyi, N. Prynada, Tet. Lett. 41 (2000) 7997.
[10] O.I. Kolodyazhnyi, N. Prinada, Russ. J. Gen. Chem. 71 (2001) 646.
[11] O.J. Scherer, R. Anselmann, Inorg. Synth. 25 (1989) 7.
[12] A. Bashall, E.L. Doyle, F. Garcia, G.T. Lawson, D.J. Linton, D. Moncrieff, M.
McPartlin, A.D. Woods, D.S. Wright, Chem. Eur. J. 8 (2002) 5723.
[13] G.M. Sheldrick, Acta Cryst. A 64 (2008) 112.
[14] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.
[15] Persistence of Vision Raytracer, Version 3.6, 2004, Retrieved from http://
www.povray.org/download/.
Acknowledgements
This work was supported by DePaul University, College of Liber-
al Arts and Sciences, Undergraduate Summer Research Program,
Undergraduate Research Assistantship Program, and Faculty Sum-
mer Research and Development grants. NMR instrumentation
[16] N. Burford, T.S. Cameron, K.D. Conroy, B. Ellis, C.L.B. MacDonald, R. Ovans, A.D.
Phillips, P.J. Ragogna, D. Walsh, Can. J. Chem. 80 (2002) 1404.
[17] F. Garcia, R.A. Kowenicki, L. Riera, D.S. Wright, Dalton Trans. (2005) 2495.
[18] T.E. Mueller, D.M.P. Mingos, Trans. Met. Chem. (London) 20 (1995) 533.