α-Zirconium phosphonates: Versatile supports for N-heterocyclic carbenes

Article abstract of DOI:10.1039/b821301a

α-Zirconium phosphonates derivatised with N-heterocyclic carbenes provide a versatile platform for organocatalysis and metal-catalysed transformations, including the ring-opening polymerisation of cyclic esters, and olefin metathesis and hydroformylations by surface-bound ruthenium, rhodium and iridium complexes. The Royal Society of Chemistry.

Full text of DOI:10.1039/b821301a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
a-Zirconium phosphonates: versatile supports for N-heterocyclic
carbenesw
Simona Chessa, Nigel J. Clayden, Manfred Bochmann* and Joseph A. Wright*
Received (in Cambridge, UK) 27th November 2008, Accepted 12th December 2008
First published as an Advance Article on the web 13th January 2009
DOI: 10.1039/b821301a
a-Zirconium phosphonates derivatised with N-heterocyclic
carbenes provide a versatile platform for organocatalysis and
metal-catalysed transformations, including the ring-opening
polymerisation of cyclic esters, and olefin metathesis and hydro-
formylations by surface-bound ruthenium, rhodium and iridium
complexes.
followed by reaction with ZrOCl2. Thus the first route involves
refluxing 2 with 1-methyl-1H-imidazole in 1,4-dioxane for two
days to give the methylimidazolium salt 3a.z Monitoring the
substitution reaction by solid-state 13C NMR spectroscopy
indicated that, remarkably, conversion of the bromobutyl
moieties was quantitative, and the reaction proceeded both
on the external and internal layer surfaces of the ZrP.
The –CH2Br signal at d 45.3 was replaced by a new signal
at d 50.0 attributed to CH2–N, accompanied by the signals
for the imidazolium ring at d 137.7, 124.1 and 122.9, and d 37.1
for the N–Me group. The presence of free bromide ions
could be verified by titration. Replacing –CH2Br
by –CH2–C3H3N2Me+Brꢀ should result in a significant expan-
sion of the interlayer spacing. The starting material has peaks
in the powder X-ray diffraction at 2y values of approximately
5.71 and 11.31 (see ESIw); these were not seen in the product.
New peaks were not evident; the limitation on low angle
measurements means that very large layer spacings cannot
be detected. On the other hand, SEM imaging showed little
change in morphology after reaction (see ESIw). Intercalation
of primary alkyl amines into zirconium phosphonates is
known,9 but can be highly selective,10 while reaction with
larger molecules has only been reported when the Zr
phosphonates have become exfoliated.11
The immobilisation of organo and metal-based catalysts
on a variety of solid supports remains crucial for many
applications.1 a-Zirconium phosphonate (ZrP) materials,
a-[Zr(O3PR)2]N, are readily accessible hybrid organic–
inorganic materials with well-defined structures and are avail-
able with high degrees of crystallinity.2 ZrPs form layer
structures that show little variation with the identity of the
R group; the lack of surface hydroxy groups can be a
significant advantage of ZrPs in the case of early transition
metal complexes.3 The structural homogeneity of these
supports facilitates their characterisation by solid-state
NMR spectroscopy and X-ray powder diffraction. We report
here on the synthesis of new ZrPs functionalised with imida-
zolium and N-heterocyclic carbene substituents, and their use
as a platform for metal complexes and catalytic applications.
Since the first isolation of free N-heterocyclic carbenes
(NHCs),4 these ligands have become a widespread tool in
catalysis.5 Immobilising NHC complexes on polymeric or
inorganic supports has been reported; many examples involve
binding an NHC-containing complex to the surface by
another tether,6 and the use of the NHC ‘‘side-arm’’ to bind
the ligand to silica has also been described.7 Combining NHCs
with the well-ordered ZrP support opens new possibilities for
well-controlled catalysis over a wide range of catalyst loadings.
The synthesis of ZrP-supported NHCs is outlined in
Scheme 1. The 4-bromobutyl phosphonate ester 1 is available
by the Arbuzov reaction of P(OPri)3 and 1,4-dibromobutane.8
Acid hydrolysis followed by hydrothermal reaction with
ZrOCl2 in the presence of HF as crystallisation aid affords
the bromobutyl zirconium phosphonate 2 in high yield.3
Imidazolium derivatisation can be achieved by two alternative
synthetic routes: ZrPs with sterically unhindered heterocycles
are best made by heating 2 with imidazoles, whereas NHCs
with bulky substituents are accessible by first synthesising the
corresponding imidazolium-substituted phosphonic acid,
In contrast to the ready formation of 3a, the analogous
reaction with aryl-substituted imidazoles C3H3N2R [R =
2,6-Pri2C6H3 (b), 2,4,6-C6H2Me3 (c)] did not yield the
desired solids, and the bromide 2 was recovered essentially
unchanged. However, aryl imidazoles reacted successfully with
the phosphonate ester 1 to yield the imidazolium salts 4b and
4c. Treatment of neat 4b with ZrOCl2 resulted only in
Wolfson Materials and Catalysis Centre, School of Chemical Sciences
and Pharmacy, University of East Anglia, Norwich, UK NR4 7TJ.
E-mail: joseph.wright@uea.ac.uk; Fax: +44 (0)1603 592003
w Electronic supplementary information (ESI) available: Synthetic
methods for all compound described, SEM images and powder
diffraction traces for 3a, 5b,c. See DOI: 10.1039/b821301a
Scheme 1 Formation of functionalised zirconium phosphonates.
Conditions: (i) aq. HBr, reflux, 48 h; (ii) ZrOCl2ꢁ8H2O/HF (1 : 2),
1 : 1 H2O–PrOH, 80 1C, 5 days; (iii) N-methylimidazole; (iv) 1,4-
dioxane, reflux; (v) N-arylimidazole. a, R = Me; b, R = 2,6-C6H3Pri2,
c, R = 2,4,6-C6H2Me3.
ꢂc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 797–799 | 797

View Article Online
Scheme 2 Applications of ZrP-supported NHCs. a, R = Me; b, R = 2,6-C6H3Pri2, c, R = 2,4,6-C6H2Me3.
dichloromethane-soluble oligomers; however, the reaction of
ZrOCl2 with a mixture 1 and 4b gave a solid of approximate com-
position Zr(O3PC4H8Br)1.4(O3PC4H8C3H3N2C6H3Pri2Br)0.6
(5b), as determined elemental analysis, i.e. a zirconium
phosphonate with mixed substituents. The IR band at
1545 cmꢀ1 and the solid-state 13C MAS-NMR spectra
confirmed the presence of the imidazolium moiety.
Zr(O3PC4H8Br)1.05(O3PC4H8C3H3N2C6H3Me3Br)0.95 (5c) was
obtained analogously.
NMR value of d 34.54 for the model compound 12. 31P
MAS-NMR spectroscopy suggests that in 15b 1 in every 8
NHCs are ligated to Ru, and 1 in every 12 for in the N-mesityl
compound 15c. Compound 15b catalysed the ring-closing
metathesis of diethyl diallylmalonate (substrate–Ru ratio
280 : 1) with quantitative conversion in 30 min. However,
both the model catalyst 12 and the supported system are
unstable under these conditions and turn brown at the end
of the reaction, accompanied by deactivation. Attempts to
generate more stable catalysts by treating 15c and 13 with
excess pyridine gave the bis-pyridine adducts 16c and 14, in
analogy to reported complexes with saturated NHC ligands.16
Both the homogeneous and ZrP-supported catalysts are
indeed more stable but less reactive, with high conversions
being reached over 24 h.
The use of NHC-derivatised ZrPs as metal supports was
explored using Ir, Rh, Ru and Au complexes. For example, the
reactions of 3a or 5b,c with KOBut gave the carbene–ZrPs 6,
which readily formed complexes with [Ir(COD)Cl]2 or
[Rh(COD)Cl]2 (Scheme 2, COD = 1,5-cyclo-octadiene).
ICP-AES indicated a rhodium loading of 0.6 mmol gꢀ1
.
Treatment of the resulting solids with carbon monoxide gave
the supported metal dicarbonyls 7 (Ir) and 8 (Rh), respectively.
IR spectroscopy and comparison with the unsupported
complexes 9–1112 demonstrated the formation of surface-
bound iridium and rhodium dicarbonyl species.
ZrP-supported gold catalysts were obtained similarly by
treating 5b,c with Ag2O followed by the addition of
AuCl(THT) in CH2Cl2 (THT = tetrahydrothiophene). A grey
solid 17b,c was obtained, the solid-state 13C MAS-NMR
spectrum of which showed a resonance at d 171.0 character-
istic of a gold-coordinated carbene. The loading level was
0.12 mmol Au per g. On activation with silver trifluoromethane-
sulfonate, 17c proved a highly selective catalyst for the inter-
molecular hydroamination of phenylacetylene with aniline;
whereas soluble (NHC)AuOTf gave a mixture of 18 and 19
in a ratio of 75 : 25 [eqn (1)], the ZrP-supported analogue gave
18 exclusively.
Rhodium complexes immobilised on solid supports have
been very successful as hydroformylation catalysts.13 The
activity of 8a in hydroformylations was tested using hex-1-
ene at 100 1C under 10 bar of CO–H2 (1 : 1). There was indeed
quantitative hex-1-ene conversion to heptanal as a 55 : 45 n/iso
mixture. However, IR studies showed that the metal could not
be recovered unchanged and had formed new, inactive species.
The immobilisation of ruthenium-based metathesis catalysts
is gaining momentum.6,14 ZrP-supported metathesis catalysts
were prepared by reacting 6b with RuCl2(PCy3)2(QCHPh)15
to give a light pink solid 15b. The solid-state 31P MAS-NMR
spectrum showed peaks at d 5.9 and 37.1 for the phosphonate
and PCy3, respectively; the latter compares well with the 31P
N-Heterocyclic carbenes are effective organocatalysts in
their own right.17 In this context, the polymerisation of cyclic
esters18 is of particular interest, and Hedrick and co-workers
have reported the successful organocatalytic polymerisation of
L-lactide and e-caprolactone.19 The supported NHC catalyst
6a was prepared from 3a and KOBut, followed by repeated
washing with THF to remove any excess tert-butoxide.
The addition of e-caprolactone in the presence of benzyl
alcohol as co-initiator led to the immediate evolution of
heat and rapid polymerisation. The reaction was quantitative
ꢂc
This journal is The Royal Society of Chemistry 2009
798 | Chem. Commun., 2009, 797–799

View Article Online
Table 1 e-Caprolactone polymerisation: recycling of catalyst (average
and C. W. Jones, J. Am. Chem. Soc., 2007, 129, 8426–8427.
Polymer supports; (d) N. E. Leadbeater and M. Marco, Chem.
Rev., 2002, 102, 3217–3274; (e) R. Haag and S. Roller, Top. Curr.
Chem., 2004, 242, 1–42; (f) Y. Uozumi, Top. Curr. Chem., 2004,
of two runs)a
ꢀ
Mn/kDa
ꢀ
Mn/Mw
ꢀ
Run
242, 77–112. Dendrimer supports; (g) C. Muller, M. G. Nijkamp
¨
1
2
3
4
5
29 000
29 000
31 500
53 000
55 000
1.8
2.0
1.9
2.0
1.8
and D. Vogt, Eur. J. Inorg. Chem., 2005, 4011–4021.
2 (a) G. Alberti, U. Costantino, S. Allulli and N. Tomassini, J. Inorg.
Nucl. Chem., 1978, 40, 1113–1117; (b) G. Alberti, M. Casciola,
U. Costantino and R. Vivani, Adv. Mater. (Weinheim, Ger.), 1996,
8, 291–303; (c) A. Clearfield and Z. Wang, J. Chem. Soc., Dalton
Trans., 2002, 2937–2947.
a
Each run: monomer–BnOH 190 : 1, 250 mg 3a, 50 cm3 THF, 30 min.
3 S. Beck, A. R. Brough and M. Bochmann, J. Mol. Catal. A: Chem.,
2004, 220, 275–284. The dehydroxylation of silica can have sig-
nificant effects on the activity of the surface-bound catalyst. See for
example: R. M. Gauvin, T. Chenal, R. A. Hassan, A. Addad and
A. Mortreux, J. Mol. Catal. A: Chem., 2006, 257, 31–40.
4 A. J. Arduengo, III, R. L. Harlow and M. Kline, J. Am. Chem.
Soc., 1991, 113, 361–363.
5 (a) F. E. Hahn and M. C. Jahnke, Angew. Chem., Int. Ed., 2008, 47,
3122–3172; (b) W. A. Herrmann, Angew. Chem., Int. Ed., 2002, 41,
1290–1309.
within 30 minutes.y The involvement of the NHC functional-
ities (as opposed to other surface or solution species) was
verified by treating 3a with KOBut, PhCH2OH and e-capro-
lactone under identical conditions: no polymerisation
took place.
The ZrP-based NHC catalyst can easily be recycled. To this
end the solid was separated by centrifugation, washed with
KOBut solution and further solvent, followed by the addition
of monomer and fresh benzyl alcohol. The catalyst retained its
activity over at least five reaction cycles (Table 1), with good
control of polymerisation across all runs and conversions
of 498% in all cases.
6 W. J. Sommer and M. Weck, Coord. Chem. Rev., 2007, 251,
860–873.
7 B. Cetinkaya, N. Gurbuz, T. Sec
A: Chem., 2002, 184, 31–38.
8 (a) A. Eberhard and F. H. Westhelmer, J. Am. Chem. Soc., 1965,
87, 253–260; (b) O. Garcıa-Barradas and E. Juaristi, Tetrahedron:
¨
¸ kin and I. Ozdemir, J. Mol. Catal.
¸
¨
¨
´
Asymmetry, 1997, 8, 1511–1514.
In summary, a-zirconium phosphonates offer much
potential for the construction of selectively functionalised
solid surfaces. By supporting N-heterocyclic carbenes on the
surface, a number of metal species can be heterogenised. The
recyclablity of the supported NHC has been demonstrated in
the polymerisation of e-caprolactone. The synthesis of mixed
zirconium phosphonates with well-isolated active surface
species is in progress.
9 See for example: G. Alberti, M. Casciola, M. A. Donnadio,
P. Piaggio, M. Pica and M. Sisani, Solid State Ionics, 2005, 176,
2893–2898.
10 T. Kijima, S. Watanabe, K. Ohe and M. Machida, J. Chem. Soc.,
Chem. Commun., 1995, 75.
11 See for example: A. Chaudhari and C. Kumar, Microporous
Mesoporous Mater., 2005, 77, 175–187.
12 See ESIw for a listing of the CRO stretches for all carbonyl
complexes reported here. For a listing of literature CRO stretch-
ing frequencies for Ir- and Rh-NHC complexes, see: M. Iglesias,
D. J. Beetstra, A. Stasch, P. N. Horton, M. B. Hursthouse,
S. J. Coles, K. J. Cavell, A. Dervisi and I. A. Fallis, Organo-
metallics, 2007, 26, 4800–4809.
13 C. P. Mehnert, R. A. Cook, N. C. Dispenziere and M. Afeworki,
J. Am. Chem. Soc., 2002, 124, 12932–12933; A. Riisager,
R. Fehrmann, M. Haumann and P. Wasserscheid, Eur. J. Inorg.
Chem., 2006, 695–706; C. P. Mehnert, Chem.–Eur. J., 2005, 11,
50–56.
14 H. Clavier, K. Grela, A. Kirschning, M. Mauduit and S. P. Nolan,
Angew. Chem., Int. Ed., 2007, 46, 6786–6801.
15 P. Schwab, R. H. Grubbs and J. W. Ziller, J. Am. Chem. Soc.,
1996, 118, 100–110.
16 J. A. Love, J. P. Morgan, T. M. Trnka and R. H. Grubbs, Angew.
Chem., Int. Ed., 2002, 41, 4035–4037.
This work was supported by the Engineering and Physical
Sciences Research Council. We thank Mr Stephen Bennett and
Mr Bertrand Leze for powder diffraction studies.
Notes and references
z The chlorobutyl analogue of 2 also reacts with alkyl imidazoles but
more slowly. It was possible to quaternise both 1-methyl-1H-imidazole
and 1-butyl-1H-imidazole with reasonable conversion, provided suffi-
ciently long reaction times were employed.
y Typical polymerisation procedure: A solution of KOBut (50 mg,
0.44 mmol) in THF (2 cm3) was added to a suspension of 3a (250 mg,
0.73 mmol) in THF (5 cm3), and the mixture stirred for 30 min. The
solid was allowed to settle, the solvent was filtered off and washed with
THF (2 ꢃ 5 cm3). The solid was resuspended in THF (50 cm3), and
PhCH2OH (0.030 cm, 0.30 mmol) was added. Addition of e-capro-
lactone (6.0 cm3, 56 mmol) led to rapid polymerisation with evolution
of heat. After 30 min, the solution was centrifuged at 3000 rpm for
30 minutes. The solution was decanted, fresh THF (30 cm3) added to
the catalyst and the mixture was centrifuged again. The combined
THF solutions were poured into MeOH to precipitate the product.
For recycling experiments, the recovered catalyst was resuspended in
THF, treated with an aliquot of KOBut followed by washing with
more THF before adding alcohol and monomer as before.
17 (a) N. Marion, S. Dıez-Gonzalez and S. P. Nolan, Angew. Chem.,
´
Int. Ed., 2007, 46, 2988–3000; (b) D. Enders, C. Grondal and M. R.
´
M. Huttl, Angew. Chem., Int. Ed., 2007, 46, 1570–1581.
¨
18 (a) R. E. Drumright, P. R. Gruber and D. E. Henton, Adv. Mater.
(Weinheim, Ger.), 2000, 12, 1841–1846; (b) O. Dechy-Cabaret,
B. Martin-Vaca and D. Bourissou, Chem. Rev., 2004, 104,
6147–6176.
19 (a) E. F. Connor, G. W. Nyce, M. Myers, A. Mock and
¨
J. L. Hedrick, J. Am. Chem. Soc., 2002, 124, 914–915;
(b) A. P. Dove, R. C. Pratt, B. G. Lohmeijer, D. A. Culkin,
E. C. Hagberg, G. W. Nyce, R. M. Waymouth and J. L. Hedrick,
Polymer, 2006, 47, 4018–4025; (c) D. A. Culkin, W. Jeong,
S. Csihony, E. D. Gomez, N. P. Balsara, J. L. Hedrick and
R. M. Waymouth, Angew. Chem., Int. Ed., 2007, 46, 2627–2630.
1 (a) Immobilized Catalysts, ed. A. Kirschning, Top. Curr. Chem.,
2004, 242. Silica supports; (b) A. Corma and H. Garcia, Adv.
Synth. Catal., 2006, 348, 1391–1412; (c) J. C. Hicks, B. A. Mullis
ꢂc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 797–799 | 799