The facile assembly of bis-, tris- and poly(triazaphosphole) systems using click chemistry

Article abstract of DOI:10.1039/c3dt50505g

Uncatalysed 1,3-dipolar cycloaddition reactions between two phosphaalkynes, PCR (R = Bu^t or Me), and a series of di-, tri- and poly-azido precursor compounds have given very high yields of a range of triazaphosphole substituted systems. These comprise the 1,1′-bis(triazaphosphole) ferrocenes, [Fe{C₅H₄(N₃PCR)}₂], the tris(triazaphosphole)cyclohexane, cis-1,3,5-C₆H₉(N ₃PCBu^t)₃, and the poly(allyltriazaphosphole)s, {C₃H₅(N₃PCR)}_∞. Electrochemical studies on the 1,1′-bis(triazaphosphole)ferrocenes reveal the compounds to undergo reversible 1-electron oxidation processes, at significantly more positive potentials than ferrocene itself. Attempts to chemically oxidise one 1,1′-bis(triazaphosphole)ferrocene with a silver salt, Ag[Al{OC(CF₃)₃}₄] were not successful, and led to the formation of a silver coordination complex, [{Fe[μ-C ₅H₄(N₃PCBu^t)]₂(μ-Ag)} ₂][Al{OC(CF₃)₃}₄]₂, thereby demonstrating the potential the reported triazaphosphole substituted systems possess as novel ligands in coordination chemistry. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3dt50505g

Dalton
Transactions
View Article Online
View Journal
PAPER
The facile assembly of bis-, tris- and poly-
(triazaphosphole) systems using “click” chemistry†
Cite this: DOI: 10.1039/c3dt50505g
Sam L. Choong,a Ayman Nafady,a,b Andreas Stasch,a Alan M. Bonda and
Cameron Jones*a
Uncatalysed 1,3-dipolar cycloaddition reactions between two phosphaalkynes, PuCR (R = But or Me),
and a series of di-, tri- and poly-azido precursor compounds have given very high yields of a range of tri-
azaphosphole substituted systems. These comprise the 1,1’-bis(triazaphosphole)ferrocenes, [Fe-
{C5H4(N3PCR)}2], the tris(triazaphosphole)cyclohexane, cis-1,3,5-C6H9(N3PCBut)3, and the poly(allyltri-
azaphosphole)s, {C3H5(N3PCR)}∞. Electrochemical studies on the 1,1’-bis(triazaphosphole)ferrocenes
reveal the compounds to undergo reversible 1-electron oxidation processes, at significantly more positive
potentials than ferrocene itself. Attempts to chemically oxidise one 1,1’-bis(triazaphosphole)ferrocene
with a silver salt, Ag[Al{OC(CF3)3}4] were not successful, and led to the formation of a silver coordination
complex, [{Fe[μ-C5H4(N3PCBut)]2(μ-Ag)}2][Al{OC(CF3)3}4]2, thereby demonstrating the potential the
reported triazaphosphole substituted systems possess as novel ligands in coordination chemistry.
Received 25th February 2013,
Accepted 21st March 2013
DOI: 10.1039/c3dt50505g
www.rsc.org/dalton
mesitylene and triethylamine,3 their electronic properties were
predicted to be very different. While N-heterocyclic carbenes
Introduction
Bi- and higher dentate P-donor ligands are ubiquitous in (NHCs) are generally accepted to be strong σ-donors and weak
coordination chemistry, and their complexes have found a π-acceptors,4 compounds such as 1 and 2 which contain
multitude of applications in, for example, catalysis, small phosphaalkene moieties, RPvCR2, are known to be weaker
molecule activations and the design of functional models of σ-donors and stronger π-acceptors.5,6 These attributes have led
enzymes.1 While numerous synthetic protocols have been to the emergence of complexes of unsaturated P-donor ligands
developed to access multi-dentate P-donor ligand systems pos- in many areas of homogeneous catalysis.6
sessing varying steric and electronic properties, the identifi-
We wished to extend our preliminary study to the prepa-
cation of novel, facile, high yielding routes to such compounds ration of a range of new polyfunctional triazaphosphole con-
is a perpetual goal of inorganic chemists. In this respect, we taining compounds, thereby demonstrating the generality of
recently reported on the rapid and essentially quantitative the synthetic route that gave 1 and 2. Considering our prior
syntheses of the tris(triazaphosphole)s, 1 and 2, via catalyst comparisons of spatially similar triazaphosphole and NHC
free “click” reactions between organotriazides and the phospha- containing compounds, in this study we chose to target both
alkynes, PuCR (R = But or Me).2 The utility of one of these 1,1′-bis(triazaphosphole)ferrocenes, and polymer supported
(1, R = But) as a tripodal P3-ligand for the stabilisation of low- poly(triazaphosphole)s, as the NHC analogues of these systems
valent platinum fragments was demonstrated in that report. have been recently studied.7,8 In addition, the pursuit of
One of the driving forces for the syntheses of 1 and 2 was the further examples of tripodal P3-ligands related to 1 and 2 was
fact that although the compounds are essentially isostructural an objective of this work.
to previously reported tris(N-heterocyclic carbene) substituted
aSchool of Chemistry, PO Box 23, Monash University, VIC, 3800, Australia.
E-mail: cameron.jones@monash.edu
bDepartment of Chemistry, College of Science and Sustainable Energy Technologies
(SET) Centre, King Saud University, PO Box 2455, Riyadh-11451, Kingdom of Saudi
Arabia
†Electronic supplementary information (ESI) available: Crystal data, details of
data collections and refinements for 3, 4, and [Fe{C5H4(N3PCButH)}2][CF3SO3]2;
ORTEP diagram for [Fe{C5H4(N3PCButH)}2][CF3SO3]2; and cyclic voltammograms
for 3 and 4. CCDC 925610–925612. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c3dt50505g
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
174.1 ppm), with the higher field resonance being consider-
ably less intense than the lower field resonance in both cases.
Results and discussion
Phosphaalkynes, PuCR, are well known to undergo uncata- It is believed this observation derives from the previously
lysed 1,3-dipolar cycloaddition reactions with organoazides, reported presence of some branching within the polymer back-
R′N3, to give 1,2,3,4-triazaphospholes, R′N3PCR, in close to bone of the poly(allylazide) precursor.11
1
quantitative yields, and with complete regioselectivity.9 This
The H and 13C{1H} NMR spectra of 3–5 are unexceptional
reactivity is closely related to copper-catalysed Huisgen alkyne– and are consistent with their proposed structures. It is of note
azide cycloadditions, and both reaction classes fulfil all of the that the signals in the spectra for 5 are relatively sharp, which
requirements of “click” chemistry.10 In order to access the pro- suggests it predominates as one isomeric form in solution.
posed targets of this study, the sterically hindered and unhin- Based on steric considerations, this is most likely the chair
dered phosphaalkynes, PuCR (R = But or Me), were reacted at form, as depicted in Scheme 1. In contrast, and not surpris-
1
ambient temperature with either 1,1′-diazidoferrocene, cis- ingly, all the resonances in the H NMR spectra for the poly-
1,3,5-triazidocyclohexane, or poly(allylazide) to give the triaza- mers, 6 and 7, are broad. Both polymers were subject to gel
phosphole compounds, 3–7, in quantitative or near quantitat- permeation chromatography which yielded average molecular
ive yields (Scheme 1). It is of note that the product from the weights of 2087 g mol−1 for 6, and 1948 g mol−1 for 7.
reaction of PuCMe with triazidocyclohexane was not stable in
X-ray crystal structure analyses of the 1,1′-bis(triazaphosp-
solution and could not be isolated. Moreover, monitoring the hole)ferrocenes, 3 and 4, were carried out, and the molecular
progress of the other reactions using 31P NMR spectroscopy, structures of the compounds are depicted in Fig. 1 and 2
revealed that the formation of 3–5 was complete within respectively. In both, the two cyclopentadienyl rings are
30 min, whereas the formation of the triazaphosphole poly- eclipsed, though in the bulkier system, 3, the triazaphosphole
mers, 6 and 7, required ca. 2 days to reach completion.
rings are not eclipsed and do not appear to have any inter-
Compounds 3–7 are thermally very stable in the solid state, action with each other. In contrast, the triazaphosphole rings
and the ferrocene and polymer based systems are indefinitely in 4 are almost perfectly eclipsed, and the distance between
stable in the air. In contrast, the triazaphosphole substituted the two centroids of those heterocycles (3.58 Å, cf. Cpcent.
–
cyclohexane, 5, is mildly air sensitive in the solid state. All of Cpcent. = 3.30 Å) indicates a degree of intramolecular π–π stack-
the molecular complexes exhibit single sharp resonances in ing. Indeed, previous computational studies have predicted
their 31P{1H} NMR spectra, which lie in a narrow range (δ = significant aromatic character to 1,2,3,4-triazaphospholes,12
173.6–179.1 ppm), and are more than 200 ppm downfield from which would account for the π-stacking behaviour in 4. Further-
the signals for the phosphaalkyne starting materials.5 The more, the heterocycle and cyclopentadienyl rings on each side
positions of the signals are, however, fully consistent with of 4 are close to co-planar (dihedral angle = 18.9° mean),
what is expected for triazaphospholes2,9 and phosphaalkenes,5 which could allow for a degree of conjugation over each
based on prior studies. It is interesting that the solution state C5H4(N3PCMe) unit. Presumably similar π-stacking between
31P{1H} NMR spectra of the two polymers each display
two broad signals (6: δ = 176.7, 171.7 ppm; 7: δ = 179.0,
Fig. 1 Molecular structure of 3 (25% thermal ellipsoids are shown; hydrogen
atoms omitted). Selected bond lengths (Å) and angles (°): P(1)–N(1) 1.6914(15),
P(1)–C(11) 1.7168(19), N(1)–N(2) 1.348(2), N(1)–C(1) 1.428(2), N(2)–N(3)
1.307(2), N(3)–C(11) 1.355(2), N(4)–N(5) 1.347(2), N(4)–C(6) 1.424(2), N(5)–N(6)
1.306(2), N(6)–C(16) 1.358(2), P(2)–N(4) 1.6956(15), P(2)–C(16) 1.7144(18),
Scheme 1 (i) Fe(C5H4N3)2, toluene; (ii) cis-1,3,5-C6H9(N3)3, CH2Cl2; (iii)
N(1)–P(1)–C(11) 86.28(8), N(4)–P(2)–C(16) 86.41(8), N(3)–C(11)–P(1) 113.08(14),
N(6)–C(16)–P(2) 112.95(13).
{C3H5(N3)}∞, CH2Cl2.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
peak-to-peak separations (ΔEp = Eopx − Epred) for both com-
pounds are similar to that known for the reversible oxidation
of ferrocene under the same experimental conditions (ΔEp
=
70 mV at 0.1 V s−1), confirms that the one-electron oxidation
processes are electrochemically reversible (fast heterogeneous
electron-transfer) at a glassy carbon electrode. The diffusion
coefficients (D) of 3 and 4 were determined in CH2Cl2/0.1 M
[NBu4][PF6] via application of the Randles–Sevcik equation.
The values obtained were 4.25 × 10−6 and 5.85 × 10−6 cm2 s−1
respectively. These values are smaller than that obtained for
ferrocene (D = 1.67 × 10−5 cm2 s−1) under similar conditions.13
This is consistent with the substitution at the Cp rings in the
triazaphosphole-based ferrocene compounds, which results in
higher molecular masses, and hence slower rates of diffusion.
The fact that the formal reversible potentials of 3 and 4 are
identical (E0f(3,4) = 0.34 V vs. Fc0/+) implies that the redox
Fig. 2 Molecular structure of 4 (25% thermal ellipsoids are shown; hydrogen
atoms omitted). Selected bond lengths (Å) and angles (°): P(1)–N(1) 1.688(2),
P(1)–C(11) 1.717(3), N(1)–N(2) 1.352(3), N(1)–C(1) 1.429(3), N(2)–N(3) 1.311(3),
N(3)–C(11) 1.364(4), N(3)–C(11) 1.364(4), N(4)–N(5) 1.362(3), N(4)–C(6) couples are not very sensitive to substituent effects (tert-butyl
1.431(3), N(5)–N(6) 1.305(3), N(6)–C(13) 1.363(4), N(1)–P(1)–C(11) 86.35(13),
vs. methyl) in the triazaphosphole rings. The positive shift in
N(4)–P(2)–C(13) 86.11(13), N(3)–C(11)–P(1) 113.1(2), N(6)–C(13)–P(2) 113.5(2).
E0f values compared to the [FeCp2]0/+ couple is attributed to an
electron-withdrawing effect of the triazaphosphole ligands,
which results in an increase in the positive charge on the iron
the heterocycles of 3 does not occur due to their greater steric
bulk. The intra-heterocycle geometries of 3 and 4 are very close
atom, thereby making its oxidation more difficult.
Oxidative bulk electrolyses undertaken inside a glovebox, at
to those reported for 1,2 and again imply a degree of aromatic 20 °C, for (1–3 mM) solutions of 3 confirmed the one-electron
delocalisation over those rings. Of most note are their P–C dis-
stoichiometry (n = 1.0 0.05 F/eq.). However, steady-state volta-
tances (ca. 1.72 Å) which lie between the normal values for mmograms obtained with a Pt-microelectrode before and after
localised double (ca. 1.66 Å) and single (ca. 1.87 Å) bonds.5
Compounds 3 and 4 have the potential to act as bidentate
bulk electrolyses (see ESI†), revealed that when 2 mM of 3 was
exhaustively electrolysed at Eappl = 0.6 V vs. [FeCp2]0/+ in
ligands towards metal fragments, MLn, or as bridging or CH2Cl2/[Bu4N][PF6] only 70% of 3+ was generated, assuming
linker ligands in 1-dimensional coordination polymers, [{Fe- equal diffusion coefficients for 3 and 3+. As the electrolysis pro-
[μ-C5H4(N3PCR)]2(μ-MLn)}∞]. If examples of the latter could be
ceeded, the solution changed colour from yellow to deep
prepared, they could act as molecular wires, by allowing elec- green. Reductive back electrolysis, in which the initially gener-
tron transfer between the redox active iron sites, the triaza- ated cation, 3+, was reduced back to the neutral starting
phosphole coordinated metal fragment, and/or the
material, 3, was carried out at Eappl = 0.2 V. The total charge
C5H4(N3PCR) rings themselves. In order to shed some light on passed was equivalent to 0.5 F/eq. and the solution colour
the redox properties of 3 and 4, their electrochemistry was became deep orange. Cyclic and near steady-state linear sweep
investigated. The cyclic voltammetric behaviour of the com-
voltammograms recorded after complete reductive electrolysis
pounds at a glassy carbon electrode in dichloromethane solu- were consistent with the regeneration of approximately 55% of
tions containing either [Bu4N][B(C6F5)4] or [Bu4N][PF6] as the 3, along with other products. One side product gave rise to a
supporting electrolyte is similar. Both compounds exhibit well-
reversible process at the same potential as that for the
defined oxidation processes that have the same formal revers- [FeCp2]0/+ couple, thus indicating the formation of ferrocene. A
ible potential (Ef0(3,4) = 0.34 V vs. Fc0/+) to generate the cationic second side product was observed at more positive potentials
products 3+ and 4+ respectively. Scanning the potential to very
(E0f = 0.6 V) than 30/+, and is probably derived from the triaza-
positive and negative potentials did not produce evidence of phosphole fragment. The combined exhaustive oxidation/
any additional processes, thereby indicating that the ligand is reduction results demonstrate that 3+ slowly decomposes on
electroinactive within the potential range available in dichloro-
time scales of ∼30 min or longer.
methane (−2 V to +2 V). Thus, the process detected at 0.34 V is
In an attempt to chemically oxidise 3 to 3+, it was treated
assigned to the metal based oxidation of FeII to FeIII in the fer- with
a
dichloromethane solution of Ag[Al{OC(CF3)3}4].
rocenyl moiety of both compounds.
However, instead of forming a ferrocenium salt, the reaction
Analysis of cyclic voltammetric data for 3 and 4 as a func- yielded the dimeric cationic coordination complex, [{Fe-
tion of scan rate (ν = 0.1 to 1 V s−1) revealed that their oxi- [μ-C5H4(N3PCBut)]2(μ-Ag)}2][Al{OC(CF3)3}4]2 8 (Scheme 2). This was
dations correspond to chemically reversible, diffusion-
confirmed by an X-ray crystal structural analysis of the salt,
controlled, one-electron processes. The shape of the voltam- though the poor quality of that crystal structure precludes its
mograms (see ESI†) implies that the salts, [3+][PF6−] and [4+]- inclusion here. With that said, it is clear from this structure
[PF6−], generated at the electrode surface are soluble and
that the silver centres are coordinated by N-atoms adjacent to
stable on the cyclic voltammetric time scale. The fact that the the intracyclic C-centres of two heterocycles. The absence of
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
atmosphere of high purity dinitrogen. THF, toluene and
hexane were distilled over molten potassium, diethyl ether was
distilled over a Na/K alloy, while dichloromethane and fluoro-
benzene were distilled over CaH2. Melting points were deter-
mined in sealed glass capillaries under dinitrogen and are
uncorrected. Mass spectra were recorded at the EPSRC
National Mass Spectrometric Service at Swansea University.
Microanalyses were carried out at the Science Centre, London
Metropolitan University, or Campbell Microanalytical, Otago.
Reproducible microanalyses could not be obtained for the
polymer products. IR spectra were recorded using a Perkin-
Elmer RX1 FT-IR spectrometer as Nujol mulls between NaCl
plates. 1H, 13C{1H}, 31P{1H} and 19F NMR spectra were
recorded on either Bruker DPX300 or AvanceIII 400 spec-
trometers and were referenced to the resonances of the solvent
used, external H3PO4 (85% solution in D2O), or external CFCl3.
UV/vis spectra were acquired using a Cary 1E spectrometer. Gel
permeation chromatography studies were carried out using a
Tosoh ecosystems 8310 instrument. PuCBut was synthesised
by a variation of Becker’s original method,15 viz. the [Li{N-
(SiMe3)2}] catalysed elimination of hexamethyldisiloxane from
Scheme 2 (i) Ag[Al{OC(CF3)3}4], dichloromethane.
any discernible 107/109Ag satellites in the 31P NMR spectrum of
8, provides further evidence for this. It might be thought that
compound 3 could not be oxidised by the silver salt because it
is more resistant to oxidation than ferrocene, as detailed
above. However, it has been reported that the redox potential
of Ag+ in dichloromethane is +0.65 V (vs. Fc),14 and therefore
the oxidation should proceed, on the basis of the electro-
chemical results for 3. One possibility as to why it doesn’t is
the fact that Ag+ is coordinated by the triazaphosphole arms of
3, which could impart a barrier to its oxidation. In another
attempt to oxidise 3, it was treated with a dichloromethane
solution of Ag[CF3SO3]. This gave a mixture of products from
which a few crystals of the protonated form of 3, viz. [Fe-
{C5H4(N3PCButH)}2][CF3SO3]2, were recovered. Although no
spectroscopic data could be obtained for this compound because
of its very low yield, its structure was confirmed by X-ray crystallo-
graphy (see ESI†). In an effort to intentionally prepare the com-
pound in higher yield by the reaction of 3 with triflic acid, only a
complex mixture of unidentified products was obtained.
a
tetraglyme solution of (Me3Si)PvC(But)(OSiMe3).16
PuCMe,17 1,1′-diazidoferrocene,18
cis-1,3,5-triazidocyclo-
hexane,19 and poly(allylazide)11 were prepared by literature pro-
cedures, Ag[Al{OC(CF3)3}4] was kindly supplied by Prof. Ingo
Krossing (Freiburg University), while all other reagents were
used as received.
Electrochemical studies. All measurements were performed
inside a Vacuum Atmospheres glovebox connected to
a
BAS100B computer-controlled electrochemical work station at
293 2 K in CH2Cl2 containing either 0.1 M [Bu4N][PF6] or
0.05 M [Bu4N][B(C6F5)4] as the supporting electrolyte. A stan-
dard three-electrode cell equipped with a 1.5 mm diameter
glassy carbon (Cypress) disk working electrode, a platinum
mesh counter electrode, and a silver wire coated with AgCl as
reference electrode, was used for voltammetric experiments.
For bulk electrolysis experiments, large platinum gauze and
platinum mesh baskets were used as the working electrode
and counter electrode, respectively. The reference electrode
was the same as that employed in the voltammetric studies. All
reported potentials are in V versus the ferrocene/ferrocenium
[FeCp2]0/+ redox couple. This was achieved by addition of ferro-
cene as an internal standard at the end of each experiment,
measurement of the reversible potential of [FeCp2]0/+ couple
vs. Ag/Ag+ by cyclic voltammetry, followed by conversion to
[FeCp2]0/+ scale.
Conclusion
In summary, regioselective, uncatalysed 1,3-dipolar cyclo-
addition reactions between two phosphaalkynes and a series
of di-, tri- and poly-azido precursor compounds have given rise
to very high yields of a range of triazaphosphole substituted
systems, which have potential for use as polydentate ligands
towards transition metal fragments. Electrochemical studies
on two 1,1′-bis(triazaphosphole) substituted ferrocenes have
shown that these compounds undergo diffusion controlled,
reversible one-electron oxidation processes, and that the com-
pounds are significantly more difficult to oxidise than ferro-
cene itself. Attempts to chemically oxidise one 1,1′-bis-
(triazaphosphole)ferrocene with a silver salt were not success-
ful, and instead led to the formation of a dimeric silver coordi-
nation complex.
Preparation of [Fe{C5H4(N3PCBut)}2] 3. Neat PuCBut
(0.288 cm3, 1.75 mmol) was added to a solution of 1,1′-diazido-
ferrocene (0.20 g, 0.77 mmol) in toluene (15 cm3) at 20 °C. The
resultant dark orange solution was allowed to stir at ambient
temperatures for 2 h, whereupon volatiles were removed
in vacuo and the crude product extracted into hexane (10 cm3).
The extract was filtered, and the filtrate placed at −30 °C over-
Experimental section
General considerations
night to yield orange crystals of
Synthetic studies. All manipulations were carried out using M.p. 125–127 °C (decomp.); H NMR (300 MHz, C6D6, 298 K):
standard Schlenk and glove box techniques under an δ = 1.42 (s, 18H, CH3), 3.77 (m, 4H, Cp–H), 4.64 (m, 4H, Cp–H);
3
(0.34 g, 96%).
1
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
3
13C NMR (75 MHz, C6D6, 300 K): δ = 31.4 (d, JPC = 7.9 Hz, 176.7, 171.7; IR ν/cm−1 (Nujol): 1708m, 1380s, 1279s, 1091m,
2
C(CH3)3), 35.3 (d, JPC = 15.6 Hz, C(CH3)3), 65.6 (Cp–C), 69.1 800m; GPC (DMF, 298 K): Mw(Av) = 2087 g mol−1
.
2
1
(Cp–C), 99.3 (d, JPC = 14 Hz, C(N3CP)), 197.8 (d, JPC = 130.0
Hz, PvC); 31P{1H} NMR (121 MHz, C6D6, 300 K): δ = 173.6; IR
v/cm−1 (Nujol): 1633w, 1377s, 1260s, 1218w, 1092s, 1019s,
Preparation of {C3H5(N3PCMe)}∞ 7. A solution of PuCMe in
diethyl ether (2 cm3, 0.6 mmol) was added to a solution of
poly(allylazide) (0.010 g, 0.1 mmol monomer equivalents) in
dichloromethane (2.5 cm3) at 20 °C. The resultant solution
was stirred for 3 days. All volatiles were subsequently removed
in vacuo, and the brown residue washed with 2 cm3 of hexane
876w, 800s; UV-vis, toluene solution (ε, L mol−1 cm−1): λmax
=
288 nm (1328), 368 nm (1395), 446 nm (520); MS (EI/70 eV),
m/z (%): 468.3 (M+, 27); anal. calc. for C20H26FeN6P2 C 51.30%,
H 5.60%, N 17.95%; found C 51.11%, H 5.58%, N 17.82%.
1
to give a pale brown solid (0.017 g, 100%); H NMR (400 MHz,
CDCl3, 298 K): δ = 1.32 (br., CH3), 0.82–2.45 (br., CH and CH2),
4.40–4.95 (br., CH2(N3PCR)); 31P{1H} NMR (121 MHz, CDCl3,
298 K): δ 179.0, 174.1; IR ν/cm−1 (Nujol): 1600w, 1377s, 1261m,
1093m, 1021m, 800m; GPC (DMF, 298 K): Mw(Av) = 1948 g
Preparation of [Fe{C5H4(N3PCMe)}2] 4. A solution of PuCMe
in diethyl ether (0.345 mmol in 1 cm3) was added to a solution
of 1,1′-diazidoferrocene (0.036 g, 0.13 mmol) in toluene
(2 cm3) at 20 °C. The resultant dark orange solution was
allowed to stir at ambient temperatures for 30 min, whereupon
volatiles were removed in vacuo and the crude product
extracted into diethyl ether (3 cm3). The extract was filtered,
and the filtrate placed at −30 °C overnight to yield orange crys-
tals of 4 (0.05 g, 99%). M.p. 131–133 °C (decomp.); 1H NMR
mol−1
.
Preparation of [{Fe[μ-C5H4(N3PCBut)]2(μ-Ag)}2][Al{OC(CF3)3}4]2
8. A solution of Ag[Al{OC(CF3)3}4] (0.12 g, 0.11 mmol) in
dichloromethane (5 cm3) was added to a solution of 3 (0.05 g,
0.11 mmol) in dichloromethane (5 cm3) at 20 °C. The resultant
solution was stirred for 2 h, all volatiles removed in vacuo, and
the residue extracted into fluorobenzene (2 cm3). Layering the
extract with hexane afforded red crystalline plates of 8 (0.13 g,
75%); M.p. 221–226 °C (decomp.); 1H NMR (300 MHz, C6D6,
3
(300 MHz, C6D6, 298 K): δ = 2.89 (d, 6H, JPH = 11.7 Hz, CH3),
3.73 (m, 4H, Cp–H), 4.68 (m, 4H, Cp–H); 13C NMR (75 MHz,
2
C6D6, 300 K): δ = 14.4 (d, JPC = 24.8 Hz), 64.5 (d, 3JPC = 5.4 Hz,
2
Cp–C), 65.6 (Cp–C), 99.3 (d, JPC = 20.0 Hz, C(N3CP)), 179.4 (d,
1JPC = 125.2 Hz, PvC); 31P{1H} NMR (121 MHz, C6D6, 300 K):
δ = 173.7; IR v/cm−1 (Nujol): 1789m, 1261w, 1096m, 1020m,
298 K): δ = 2.22 (s, 36H, C(CH3)3), 4.12 (m, 8H, Cp–H), 4.74 (m,
3
8H, Cp–H); 13C NMR (75 MHz, C6D6, 300 K): δ = 28.3 (d, JPC
=
8.2 Hz, C(CH3)3), 36.1 (d, 2JPC = 16.2 Hz, C(CH3)3), 65.2 (Cp–C),
836m, 799m; UV-vis, toluene solution (ε, L mol−1 cm−1): λmax
=
2
1
69.1 (Cp–C), 106.2 (d, JPC = 14 Hz, C(N3CP)), 212.2 (d, JPC
=
286 nm (1304), 368 nm (1676), 446 nm (2026); MS (EI/70 eV),
m/z (%): 384.2 (M+, 100); anal. calc. for C14H14FeN6P2
129.0 Hz, PvC); 31P NMR (121 MHz, C6D6, 300 K): δ = 163.8;
19F NMR (377 MHz, CD2Cl2, 298 K): δ = −75.7 (s); IR v/cm−1
(Nujol): 1718w, 1377s, 1261w, 1090m, 1020m, 800m; MS (EI
C
43.78%, H 3.67%, N 21.88%; found C 43.56%, H 3.61%, N
21.67%.
+
70 eV), m/z (%): 1152.2 ({Fe[μ-C5H4(N3PCBut)]2(μ-Ag)}2 , 15),
Preparation of cis-1,3,5-C6H9(N3PCBut)3 5. Neat PuCBut
(0.50 cm3, 3.0 mmol) was added to a solution of cis-1,3,5-tri-
azidohexane (0.20 g, 0.80 mmol) in dichloromethane–THF
(50 : 50 v/v, 20 cm3) at 20 °C. The reaction solution stirred for
4 h before all volatiles were removed in vacuo. The colourless
residue was extracted into a toluene–hexane mixture and
stored at −30 °C to yield colourless needle-like crystals of 5
468.3 (Fe{C5H4(N3PCBut)}2 , 100).
+
X-Ray crystallography
Crystals of 3, 4, and [Fe{C5H4(N3PCButH)}2][CF3SO3]2 suitable
for X-ray structural determination were mounted in silicone
oil. Crystallographic measurements were made using an Enraf
Nonius Kappa CCD diffractometer using a graphite monochro-
mator with Mo Kα radiation (λ = 0.71073 Å). The structures
were solved by direct methods and refined on F2 by full matrix
least squares (SHELX97)20 using all unique data. Hydrogen
atoms have been included in calculated positions (riding
model) for all structures, except the N–H protons of [Fe-
{C5H4(N3PCButH)}2][CF3SO3]2, the positional and thermal
parameters of which were freely refined. Crystal data, details of
data collections and refinement are given in Table S1 in the
ESI.†
1
(0.38 g, 93%). M.p. 224–227 °C (decomp.); H NMR (400 MHz,
CDCl3, 298 K): δ = 1.45 (s, 27H, C(CH3)3), 2.74 (m, 3H, CH2),
3.02 (m, 3H, CH2), 5.18 (m, 3H, Cy–CH); 13C{1H} NMR
2
(75 MHz, CDCl3, 298 K): δ = 31.6 (C(CH3)3), 35.6 (d, JPC = 16.2
Hz, C(CH3)3), 43.3 (CH2), 58.7 (d, 2JPC = 12.0 Hz, CH), 199.4 (d,
1JPC = 108.1 Hz, PvC); 31P{1H} NMR (121 MHz, CDCl3, 298 K):
δ 179.1; IR ν/cm−1 (Nujol): 1655w, 1205m, 1181m, 1125s,
1058s, 893m, 800m; MS (EI/70 eV), m/z (%): 507.5 (M+, 54);
anal. calc. for C21H36N9P3 C 49.70%, H 7.15%, N 24.84%;
found C 49.01%, H 7.00%, N 24.52%.
Preparation of {C3H5(N3PCBut)}∞ 6. Neat PuCBut (0.10 cm3,
0.6 mmol) was added to a solution of poly(allylazide) (0.010 g,
0.1 mmol monomer equivalents) in dichloromethane (5 cm3)
Acknowledgements
at 20 °C. The resultant solution was stirred for 3 days. All vola- CJ thanks the Australian Research Council. The EPSRC Mass
tiles were subsequently removed in vacuo, and the brown Spectrometry Service at Swansea University is also thanked. AN
residue washed with 2 cm3 of hexane to give a pale brown thanks King Saud University, Deanship of Scientific Research,
solid (0.011 g, 66%); 1H NMR (400 MHz, CDCl3, 298 K): δ = College of Science Research Center. Dr Nils Trapp (Freiburg
1.40 (br., C(CH3)3), 0.80–2.40 (br., CH and CH2), 4.50–5.00 (br., University) is thanked for his efforts refining the X-ray crystal
CH2(N3PCR)); 31P{1H} NMR (121 MHz, CDCl3, 298 K): δ = structure of compound 8.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
9 W. Rösch, T. Facklam and M. Regitz, Tetrahedron, 1987, 43,
3247.
References
1 See for example: (a) Phosphorus(III) Ligands in Homogeneous 10 (a) C. R. Becer, R. Hoogenboom and U. S. Schubert, Angew.
Catalysis, Design and Synthesis, ed. P. C. J. Kamer and
P. W. N. M. van Leeuwen, Wiley, Chichester, 2012;
Chem., Int. Ed., 2009, 48, 4900; (b) H. C. Kolb, M. G. Finn
and K. B. Sharpless, Angew. Chem., Int. Ed., 2001, 40, 2004.
(b) H. A. Mayer and W. C. Kaska, Chem. Rev., 1994, 94, 11 B. Gaur, B. Lochab, V. Choudhary and I. K. Varma,
1239; (c) C. T. Saouma and J. C. Peters, Coord. Chem. Rev.,
2011, 7–8, 920.
2 S. L. Choong, C. Jones and A. Stasch, Dalton Trans., 2010,
39, 5774.
J. Therm. Anal. Calorim., 2003, 71, 467.
12 L. Nyulászi, T. Veszprémi, J. Réffy, B. Burkhardt and
M. Regitz, J. Am. Chem. Soc., 1992, 114, 9080.
13 P. Zanello, Inorganic Electrochemistry: Theory, Practice and
Application, Royal Society of Chemistry, Cambridge, 2003.
3 (a) K. Meyer and S. C. Bart, Adv. Inorg. Chem., 2008, 60, 1;
(b) X. Hu and K. Meyer, J. Organomet. Chem., 2005, 690, 14 N. G. Connelly and W. E. Geiger, Chem. Rev., 1996, 96,
5474; (c) H. Nakai, Y. Tang, P. Gantzel and K. Meyer, Chem.
Commun., 2003, 24.
4 See for example: (a) F. E. Hahn and M. C. Jahnke, Angew.
877.
15 G. Becker, G. Gresser and W. Uhl, Z. Naturforsch., B: Chem.
Sci., 1981, 36, 16.
Chem., Int. Ed., 2008, 47, 3122; (b) W. A. Herrmann, Angew. 16 M. D. Francis, PhD thesis, University of Wales, Swansea,
Chem., Int. Ed., 2002, 41, 1290. 1998.
5 K. B. Dillon, F. Mathey and J. F. Nixon, in Phosphorus: The 17 J.-C. Guillemin, T. Janati and J.-M. Denis, J. Org. Chem.,
Carbon Copy, Wiley, Chichester, 1998.
6 P. Le Floch, Coord. Chem. Rev., 2006, 250, 627.
7 See for example: C. D. Varado Jr., V. M. Lynch and
C. Bielawski, Dalton Trans., 2009, 7253.
8 See for example: Y. He and C. Cai, Chem. Commun., 2011,
47, 12319.
2001, 66, 7864.
18 A. Shafir, M. P. Power, G. D. Whitener and J. Arnold, Orga-
nometallics, 2000, 19, 3978.
19 J. E. Bollinger, J. T. Mague, W. A. Banks, A. J. Kastin and
D. M. Roundhill, Inorg. Chem., 1995, 34, 2143.
20 G. M. Sheldrick, SHELX-97, University of Göttingen, 1997.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013