Novel organosoluble filiform zirconium phosphonates with a layered mesoporous backbone

Article abstract of DOI:10.1039/b816542d

A novel type of organosoluble and filiform zirconium phosphonate with a layered mesoporous backbone functionalized with hydroxyl and amino groups was prepared by the reaction of a functionalized phosphoric acid with ZrOCl ₂·8H₂O. The

Full text of DOI:10.1039/b816542d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Novel organosoluble filiform zirconium phosphonates with a layered
mesoporous backbone†
*
Xuebing Ma, Jing Liu, Linshan Xiao, Rui Chen, Jinqin Zhou and Xiangkai Fu
Received 24th September 2008, Accepted 14th November 2008
First published as an Advance Article on the web 5th January 2009
DOI: 10.1039/b816542d
A novel type of organosoluble and filiform zirconium phosphonate with a layered mesoporous
backbone functionalized with hydroxyl and amino groups was prepared by the reaction of
a functionalized phosphoric acid with ZrOCl2$8H2O. These filiform zirconium phosphonates has good
solubility (0.21–0.71 g mLꢀ1) in certain organic solvents (such as toluene, chloroform, tetrahydrofuran
and ethyl acetate), are immiscible in some other solvents (hexane, cyclohexane, petroleum ether and
ethanol), and can be recovered from organic solvents in high yield (95–100%) by precipitation.
zirconium phosphonates have continued to be of considerable
Introduction
interest in recent years, but unfortunately they are immiscible in
organic solvents. In this paper, by the choice of organic pendant
groups bonded to the backbone of a layered and porous zirco-
nium phosphate, we have developed the first example of an
organosoluble and filiform zirconium phosphonate (Fig. 1). This
catalyst is soluble in toluene, chloroform, tetrahydrofuran and
ethyl acetate, and immiscible in hexane, cyclohexane, petroleum
ether and ethanol. As an ideal scaffold for the immobilization of
expensive metals such as Pt, Rh and Pd, these novel zirconium
phosphonates 1–4 functionalized by amino and hydroxyl
complex groups will fill a gap in the field of homogeneous
catalysis, and combine the individual advantages of the orga-
nosolubility of organic polymers and the rigid and porous
properties of inorganic solids. Furthermore, the zirconium
phosphonates 1–4 are readily recyclable by precipitation.8
The immobilisation of homogeneous catalysts is usually used to
facilitate the separation of relatively expensive catalysts from
reaction mixtures.1 A successful homogeneous catalyst immobi-
lized on insoluble inorganic solid supports such as silica gel or
zeolites can positively influence the catalytic performance; for
example, the catalytic activity, chemoselectivity and enantiose-
lectivity can increase due to the site isolation effect,2 the confine-
ment effect,3 and the cooperative effect of the support backbone.4
Unfortunately, due to the problem of mass transfer, most exam-
ples of catalysts supported on inorganic materials have tended to
have inferior catalytic properties compared to their homogeneous
counterparts. Therefore, a new strategy, using soluble polymers
and dendrimers as supports in catalysis,5 has recently been high-
lighted as a new option for the immobilization of homogeneous
catalysts. However, owing to the flexibility of soluble polymers
and dendrimers with linear and cross-linked structures, these
catalysts can become entangled or twisted, shielding their active
catalytic centers and lowering their catalytic properties.
Experimental
General
Due to applications in the field of catalysis,6 adsorption, ion
exchange and/or functional materials,7 layered and porous
Infrared spectra were recorded on a Spectrum GX using poly-
styrene as a standard (KBr pellet). TG analysis was performed
on a SBTQ600 Thermal Analyzer (USA) with a heating rate of
Fig. 1 The structures of the organosoluble filiform zirconium phos-
phonates 1–4.
College of Chemistry and Chemical Engineering, Southwest University,
Chongqing, 400715, P. R. China. E-mail: zcj123@swu.edu.cn; Fax: +86
23 68253237; Tel: +86 23 6825 4963
† Electronic supplementary information (ESI) available: NMR and IR
spectra, and TEM images. See DOI: 10.1039/b816542d
Fig. 2 The synthesis of zirconium phosphonates 1–4.
This journal is ª The Royal Society of Chemistry 2009
1098 | J. Mater. Chem., 2009, 19, 1098–1104

View Article Online
20 C minꢀ1 from room temperature to 1000 C under flowing
vmax,/cmꢀ1 3417w (O–H), 3063, 3032s (Ph–H), 2983–2728s (CH3,
CH2, CH), 1601s, 1545s, 1492s (Ph), 1446w, 1387s (CH3, CH2,
CH), 1031vs (PO3). Found: C, 71.19; H, 7.40; N, 3.16. Anal.
Calcd for C26H32NO5: C, 71.20; H, 7.37; N, 3.19%.
ꢁ
ꢁ
compressed N2 (100 mL minꢀ1). H, 13C and 31P NMR were
1
performed on a Bruker AV-300 NMR instrument at ambient
temperature at 300, 75 and 121 MHz, respectively. All of the
chemical shifts were reported downfield in ppm relative to the
hydrogen, carbon and phosphorus resonance of TMS, chloro-
form-d1 and H3PO4 (85%) respectively. The interlayer spacings
were obtained on a DX-1000 automated X-ray power diffrac-
tometer, using Cu Ka radiation and internal silicon powder
standard with all samples. The patterns were generally measured
between 2.00ꢁ and 35.00ꢁ (2q) with a step size of 1ꢁ minꢀ1 and
X-ray tube settings of 40 kV and 2.5 mA. C, H and N elemental
analysis was obtained from an EATM 1112 automatic elemental
analyzer instrument (Thermo, USA). The size and morphology
of as-synthesized samples were determined by a Hitachi model
H-800 transmission electron microscope. N2 adsorption–
desorption analysis was carried out at 77 K on an Autosorb-1
apparatus (Quantachrome). The surface areas of zirconium
phosphonates were determined by using the BET equation, and
the pore diameters were estimated according to the BJH model.
All materials and reagents used were of analytical grade
(Astara) and were used as supplied without further purification.
The zirconium phosphonates 1–4 were prepared following the
synthetic route described in Fig. 2.
Phosphonate S-2. Mp 141–142 ꢁC (recrystallized from ethyl
3
acetate). dH (300 MHz, CDCl3, Me4Si) 7.69 (2 H, d, Ph, J1,3
¼
7.23 Hz), 7.57–7.10 (13 H, m, Ph), 5.09 (1 H, s, OCH2), 4.43 (1 H,
s, OCH2), 4.15 (1 H, t, CH), 3.89–3.76 (4 H, m, OCH2), 2.72–2.56
(2 H, m, CH2P, AB system), 2.46–2.38 (2 H, m, CH2), 1.24–1.12
(3 H, m, CH3). dC (300 MHz, CDCl3) 145.2, 142.6, 139.6, 129.2,
128.4, 128.1, 127.8, 127.3, 126.8, 126.7, 126.2, 125.9 (Ph), 88.2
3
(COH), 85.7 (CH2OH), 73.9 (NCH, d, Jcp ¼ 15.5 Hz), 62.2
(OCH2, d, 3Jcp ¼ 43.2 Hz), 50.8 (CH2P, d, 1Jcp ¼ 165.6 Hz), 38.3
3
(CH2), 16.3 (CH3, t, Jcp ¼ 6.2 Hz). dP (300 MHz, CDCl3, 85%
H3PO4) 24.4 (t). IR (KBr): vmax/cmꢀ1 3483s (O–H), 3331vs,
3285vs, 3244vs, 3170vs (Ph), 2940s, 2886s (CH3, CH2, CH),
1631s, 1599s (Ph), 1450w, 1383vs (CH3, CH2, CH), 1025vs (PO3).
Found: C, 71.63; H, 7.63; N, 3.08. Anal. Calcd for C27H34NO5:
C, 71.65; H, 7.59; N, 3.10%.
Phosphonate S-3. Mp 138–139 ꢁC. dH (300 MHz, CDCl3,
Me4Si) 7.66 (2 H, d, Ph,, 3J1,3 ¼ 6.0 Hz), 7.39–7.21 (3 H, m, Ph),
7.13–7.04 (10 H, m, Ph), 7.18–7.01 (Ph, 10 H, m), 5.39 (1 H, s,
NCH), 4.49 (2 H, d, PCH2N, 2JHP ¼ 48.0 Hz), 4.13–3.96 (4 H, m,
OCH2), 2.96–2.60 (3 H, m, NCH3), 1.30–1.19 (6 H, m, CH3). dC
(75 MHz, CDCl3) 146.5, 145.5, 144.9, 129.8, 127.9, 127.6, 127.4,
127.1, 126.8, 125.9, 125.6, 125.5 (Ph), 85.7 (C–O), 73.7 (CHN),
63.3 (OCH2, d, 2Jcp ¼ 6.0 Hz), 61.7 (NCH3, d, 3Jcp ¼ 6.0 Hz), 46.6
Synthesis of phosphonates S-1–4
To a mixture of 2 mL of THF, 4 mL of H2SO4 (25%) and
formaldehyde (2.0 g, 24.7 mmol, 37%) was added dropwise
40 mL of THF solution containing b-aminoalcohol or N-methyl-
b-aminoalcohol (10 mmol) over 30 min. This was stirred for 1 h
at 40 ꢁC, and concentrated under the reduced pressure. The
residue was charged with diethyl phosphite (1.66 g, 12 mmol) and
cationic exchange resin (0.5 g). The reaction mixture was stirred
at 60 ꢁC for 6 h, and the cationic exchange resin was removed by
filtration. For the phosphonates starting from b-aminoalcohol
(R00 ¼ H), the reaction mixture was evaporated under reduced
pressure and 10 mL of ethyl acetate was added. White needles
of the phosphonates S-1 (3.7 g, 82%) anꢁd S-2 (3.4 g, 76%) were
obtained by filtration and drying at 60 C under reduced pres-
sure. Otherwise, for the phosphonates starting from N-methyl-b-
aminoalcohol (R00 ¼ CH3), the reaction mixture was evaporated
under the reduced pressure to give a yellow oily liquid. The white
phosphonates S-3 (3.5 g, 78%) and S-4 (3.7 g, 79%) were purified
by silica gel column chromatography using ethyl acetate/petro-
leum ether (60–90 ꢁC) (v/v ¼ 1:10), evaporation and drying under
1
(PCH2, d, Jcp ¼ 100.5 Hz). dP (121 MHz, CDCl3, 85% H3PO4)
18.3 (s). IR (KBr): vmax/cmꢀ1 3286s (O–H), 3084w, 3055w, 3025s
(Ph), 2964w, 2936 w, 2884w, 2796w (CH3, CH2, CH), 1601s,
1492s (Ph), 1060 vs (PO3). Found: C, 73.84; H, 7.72; N, 3.24.
Anal. Calcd for C26H32NO4: C, 73.90; H, 7.65; N, 3.32%.
Phosphonate S-4. Mp 144–145 ꢁC. dH (300 MHz, CDCl3,
Me4Si) 7.66–7.57 (4 H, dd, Ph,), 7.34–7.08 (11 H, m, Ph), 6.19
(–OH, s), 4.16—3.97 (2 H, m, OCH2), 3,71 (1 H, t, NCH, 3J1,3
¼
18.6 Hz), 3.37–3.26 (2 H, m, OCH2), 2.97, 2.65–2.47 (2 H, dd,
PhCH2), 2.21 (3 H, s, NCH3), 1.26 (3 H, t, CH2CH3, 3J1,3 ¼ 7.0
Hz), 0.96 (3 H, t, CH2CH3, 3J1,3 ¼ 7.0 Hz). dC (75 MHz, CDCl3)
147.7, 145.0, 140.4, 129.7, 128.3, 128.0, 128.0, 126.8, 125.9, 125.8,
125.2, 125.2 (Ph), 81.5 (COH), 72.3 (CHN, d, 3Jcp ¼ 9.0 Hz), 62.2
2
1
(OCH2, d, Jcp ¼ 137.3 Hz), 48.31 (PCH2, d, Jcp ¼ 171.8 Hz),
3
47.4 (NCH3), 31.8 (CH2), 16.3 (CH3, t, Jcp ¼ 5.1 Hz). dP (121
MHz, CDCl3, 85% H3PO4) 28.5. IR (KBr): vmax/cmꢀ1 3420vs (O–
H), 3171s, 3119vs (Ph) 2982w, 2947w, 2892s (CH3, CH2, CH),
1641vs (Ph), 1117s (PO3). Found: C, 74.21; H, 7.92; N, 3.20.
Anal. Calcd for C27H34NO4: C, 74.28; H, 7.87; N, 3.21%.
ꢁ
reduced pressure at 60 C for 10 h.
Phosphonate S-1. Mp 136–137 ꢁC (recrystallized from ethyl
3
acetate). dH (300 MHz, CDCl3, Me4Si) 7.63 (2 H, d, Ph, J1,3
7.56 Hz), 7.35 (2 H, d, Ph, 3J1,3 ¼ 7.7 Hz), 7.25 (1 H, d, Ph, 3J1,3
¼
¼
Preparation of zirconium phosphonates (1–4)
7.26 Hz), 7.11–7.06 (3 H, m, Ph), 7.02–6.99 (7 H, Ph, m), 5.37
(1 H, s, OCH2), 4.52 (1 H, s, OCH2), 4.38 (1 H, s, CH), 4.08–3.93
(4 H, m, OCH2), 2.93–2.55 (2 H, m, CH2P, AB system), 1.31–1.04
(6 H, m, CH3). dC (300 MHz, CDCl3) 145.1, 142.4, 136.8, 130.0,
128.2, 127.8, 127.5, 127.3, 127.0, 126.7, 126.6 (Ph), 89.6 (C–OH,
A mixture of the phosphonate S-1 (4.4 g, 10 mmol), 50 mL of
acetic acid and 10 mL of hydrochloric acid (36%) was stirred at
80 ꢁC for 8 h. Zirconium oxychloride (3.4 g, 10 mmol) in 10 mL
of deionized water was added dropwise and stirred continued at
80 ꢁC for 4 h. The white solid zirconium phosphonate 1 was
filtered, washed with sodium carbonate (0.1 mol Lꢀ1) and water
to achieve pH ¼ 6–7, and then washed with ethanol (30 mL ꢂ 3)
2
s), 85.9 (CH2OH, s), 77.5 (NCH, s), 62.3 (OCH2, d, Jcp ¼ 46.6
Hz), 46.0 (CH2P, d, 1Jcp ¼ 165.2 Hz), 16.4 (CH3, t, 3Jcp ¼ 6.4 Hz).
dP (300 MHz, CDCl3, 85% H3PO4) 24.8–24.6 (m). IR (KBr):
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 1098–1104 | 1099

View Article Online
and dried at 60 ꢁC for 24 h under reduced pressure. White
zirconium phosphonates 2–4 were prepared according to the
same procedure.
about 1030 cmꢀ1 were caused by the Zr–O–P stretching vibra-
tion. From this data, it was concluded that functional organic
moieties containing a b-aminoalcohol were anchored on the
frame of zirconium phosphonate, whose solution 31P NMR had
a resonance in the range ꢀ10 to +20 ppm.
Zirconium phosphonate 1. 7.6 g in 90% yield. dP (121 MHz,
CDCl3, 85% H3PO4) 12.3 (s, bound). IR (KBr): vmax/cmꢀ1 3407s
(OH), 3060s, 3028s (Ph), 2820w, 2756w (CH3, CH2, CH), 1681s,
1596s, 1493s (Ph), 1448s, 1358w (CH3, CH2, CH), 1029vs
(PO3Zr). Found: C, 47.20; H, 4.48%; N, 2.14. Anal. Calcd for
Thermal gravimetric analysis. From the TG curves shown in
Fig. 3, it can be deduced that 1–4 lost 2.2–2.5 molar equivalents
of surface-bound or intercalated water below 200 ꢁC, and the
process, as usual, was endothermic. The sharp weight loss in the
temperature range of 200–500 ꢁC, accompanied by endothermic
peaks in the DSC curves, corresponded to the decomposition of
the appended organic fragments. After the volatilization of
organic fragments, zirconium phosphonates 1–4 were converted
to zirconium phosphate. In the same temperature range, the
losses of pendant organic moieties by TG analysis were 62.77,
61.75, 61.99 and 61.24 wt% respectively, which were 1.8–3.4%
wt% higher than that of the theoretical values; this can probably
be attributed to the simultaneous dehydration of the internal and
external free hydroxyl groups. Finally, the small weight losses
C24.42H31.0 N1.11O9.73P1.11Zr: C, 47.22; H, 4.94; N, 2.50%.
Zirconium phosphonate 2. 7.9 g in 92% yield. dP (121 MHz,
CDCl3, 85% H3PO4) 12.9 (s, bound). IR (KBr): vmax/cmꢀ1 3409s
(–OH), 3299s (Ph), 2956s, 2887s (CH3, CH2, CH), 1631s, 1587s
(Ph), 1382vs (CH3, CH2, CH), 1029s (PO3Zr). Found: C, 47.36;
H, 4.78; N, 2.30. Anal. Calcd for C24.84H32.76 N1.08O9.74P1.08Zr:
C, 47.54; H, 5.27; N, 2.41%.
Zirconium phosphonate 3. 7.2 g in 88% yield. dP (121 MHz,
CDCl3, 85% H3PO4) 13.44 (s, bound). IR (KBr): vmax/cmꢀ1 3408s
(OH), 3060s, 3026s (Ph), 2820w, 2797w (CH3, CH2, CH), 1681s,
1597s, 1493s (Ph), 999vs (PO3Zr). Found: C, 47.26; H, 4.67; N,
2.23. Anal. Calcd for C22.22H28.60N1.01O8.22 P1.01Zr: C, 47.33; H,
5.12; N, 2.51%.
ꢁ
ꢁ
between 800 or 900 C and 1200 C were due to phase changes
from layered to cubic ZrP2O7, as confirmed by XRD.9
Powder XRD. The interlayer distances or d-spacings of zirco-
nium planes for crystalline zirconium phosphonates can be
determined from the 00n peaks in the powder XRD pattern (via
the Bragg equation, nl ¼ 2d sinq). The d-spacings for 1–4 are
given in Fig. 4. As can be seen from Fig. 4, except for zirconium
phosphonate 3, the other XRD patterns displayed a broad 001
peak (the lowest-angle diffraction peak in the pattern), without
higher-order 00n peaks at larger angles and lower intensities.
Assuming a perpendicular orientation of the organic chains
and the bimolecular layers of b-aminoalcohol moieties with an
‘‘end-to-end’’ structure between neighbouring zirconium planes,
the predicted d-spacings of the zirconium phosphonates are
obtained by summing the zirconium phosphate layer thickness
Zirconium phosphonate 4. 7.3 g in 88% yield. dP (121 MHz,
CDCl3, 85% H3PO4) 11.57 (s, bound). IR (KBr): vmax/cmꢀ1 3408s
(OH), 3060s, 3028s (Ph), 2828w, 2795w (CH3, CH2, CH), 1681s,
1597s, 1493s (Ph), 997vs (PO3Zr). Found: C, 47.31; H, 4.46; N,
2.22. Anal. Calcd for C21.85H29.30N0.95O8.10 P0.95Zr: C, 47.24; H,
5.02; N, 2.40%.
Results and discussion
Chemical and physical characterization
10
˚
Chemical analysis. In a sealed plastic container, a mixture of
2 mL of deionized water and 50 mg of zirconium phosphonate
1–4 was treated with 2 mL of hydrofluoric acid (30%) and stirred
at room temperature for 6 h. Due to the formation of the
complex ZrF62ꢀ, 1–4 decomposed. In the resulting solution, the
zirconium(IV) and phosphonic acid content was determined by
colorimetry and 31P NMR respectively. The molar ratio of the
appended organic moieties to zirconium(IV) in 1–4 were 1.11,
1.08, 1.01 and 0.95 respectively. The empirical formulae of 1–4
were ascertained by elemental analysis and TG analysis, and are
shown in Fig. 1.
(3.2 A), twice the normal distance (projection along the P–C
bond axis) from the phosphorus atom to the farthest hydrogen
atom in the organic functional group, and twice the hydrogen
11
˚
atomic radius (0.35 A). Therefore, the calculated interlayer
spacings of 1–4 by the MM2 method are 22.1, 21.1, 19.9 and
˚
18.9 A respectively. The calculated interlayer spacings of zirco-
nium phosphonates 1–3 are in good agreement with the experi-
˚
mental values, with a just a small deviation (below 0.6 A). This
small deviation is most likely due to the two assumptions
inherent in this simple model: (i) that the organic groups are
perpendicular to the plane containing the zirconium atoms; and
(ii) that the contact between the organic groups in adjacent layers
is ‘‘end-to-end’’. As an example, the idealized chain geometry of
zirconium phosphonates 1 is shown in Fig. 5. In the idealized
model, the hydroxyl groups in the organic moieties form
hydrogen bonds with each other (verified by the strong broad
infrared absorption band), making the zirconium atoms
self-assemble in the same plane.
Infrared spectroscopy. In order to ascertain the attachment of
the functional organic moieties on the backbone of zirconium
phosphonate, infrared spectra of 1–4 were recorded in the range
4000–400 cmꢀ1. The strong and wide absorption band extending
from 3700 to 2500 cmꢀ1 and centered at 3407 cmꢀ1 was assigned
to the –O–H stretching vibration, which verified the presence of
a hydrogen bond between the hydroxyl groups on the internal
and external surfaces of the zirconium layers. The sharp
absorption bands at 1680–1450 cmꢀ1 and 700–670 cmꢀ1 were
respectively attributed to the characteristic absorptions and the
flexing vibration of the phenyl group. The absorption bands at
However, for zirconium phosphonate 4 there was a large
˚
deviation between the actual interlayer spacing of 14.18 A and
˚
the predicted interlayer spacing of 18.9 A by MM2 based on the
assumption of an ‘‘end-to-end’’ structure. This is probably
because the surface hydroxyl groups on the zirconium plane
1100 | J. Mater. Chem., 2009, 19, 1098–1104
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Fig. 3 The TG curves of the zirconium phosphonates 1–4.
Fig. 5 The idealized structure of zirconium phosphonate 1.
Fig. 4 X-Ray diffraction patterns displaying the 00n peaks of zirconium
phosphonates 1–4.
observations from the Rietveld refinement of the structure of
zirconium bis(phenylphosphonate) and zirconium p-amino-
benzylphosphonates both show that the P–C bond is not normal
to the layer and is in fact tilted by approximately 30ꢁ.12 In the
light of the possibility of forming hydrogen bonds between
the hydroxyl groups, the idealized structure of 4 was refined by
the MM2 method (see Fig. 6), although the actual structure most
increased in number with the decrease of x values from 1.11 to
0.95, providing enough large spaces to satisfy the different
structural possibilities for the arrangement of organic moieties
between the neighbouring zirconium planes. Moreover, in
regards to the first assumption mentioned above, experimental
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 1098–1104 | 1101

View Article Online
for solid-state 1–4 were found (9.4–12.3 m2 gꢀ1). The nitrogen
adsorption–desorption isotherm of zirconium phosphonate 2
(Fig. 7) was type IV, with a sharp increase in N2 adsorption at
higher P/P0 values (ꢃ0.9) and a distinct hysteresis loop (type H1),
with an almost parallel adsorption and desorption.
For zirconium phosphonates 1, 3 and 4, a distinct hysteresis
loop (type H3) without parallel or overlapped nitrogen adsorp-
tion–desorption isotherms was observed. Based on the desorp-
tion isotherm, BJH analysis gave a broad and non-uniform
distribution of pore size (in the range 1.5–5 nm) for 1–4. The
physical properties of 1–4 are listed in Table 1.
Fig. 6 The idealized structure of zirconium phosphonate 4.
Surface morphology. Generally, zirconium phosphonates
are spheroid or featureless and submicron in size with a layered
crystalline structure.13,6d–f The as-synthesized samples of the
layered mesoporous zirconium phosphonates 1–4 were dispersed
in cyclohexane, stirred for 2 h and their morphology determined
by transmission electron microscopy. Surprisingly, we saw
a novel type of filiform structure (Fig. 8).
likely includes other conformations not depicted in Fig. 5 and
Fig. 6. The calculated interlayer spacing of the refined model
˚
shown in Fig. 6 is 14.5 A, slightly higher than the detected value
˚
(14.18 A), with a small deviation (0.32 A).
˚
Nitrogen adsorption–desorption isotherms. Following the BET
treatment of the isotherms, low specific surface areas were found
Fig. 7 The nitrogen adsorption–desorption isotherm and pore distri-
bution of zirconium phosphonate 2.
Fig. 8 TEM images of the layered mesoporous zirconium phosphonates
1–4.
Table 1 The physical properties of zirconium phosphonates 1–4
Organic
moieties (%)a
Elemental analysis
Phosphonate
Calc.
Found
Calc.
59.4
Found
Surface area (m2/g)
12.3
Pore volume (ꢂ 10ꢀ2 cm3/g)
Average pore diameter (nm)
3.02
1
C
H
N
C
H
N
C
H
N
C
H
N
47.22
4.94
2.50
47.54
5.27
2.41
47.33
5.12
2.51
47.24
5.02
2.40
47.20
4.48
2.14
47.36
4.78
2.30
47.26
4.67
2.23
47.31
4.46
2.22
62.8
61.8
62.0
61.2
9.31
2
3
4
59.6
59.1
59.4
11.0
10.7
9.4
5.34
6.63
4.45
1.94
2.47
1.90
a
From thermogravimetric analysis.
1102 | J. Mater. Chem., 2009, 19, 1098–1104
This journal is ª The Royal Society of Chemistry 2009

View Article Online
From the TEM images, it can be seen that the filiform struc-
tures were produced with widths of 20–30 nm and lengths up to
hundreds of micrometers. Furthermore, according to the XPRD
analysis, the zirconium phosphonates 1–4 were layered meso-
porous materials with an interlayer spacing of about 2 nm
(except for zirconium phosphonate 4); see Fig. 4, 5 and 6. As an
example, based on the interlayer spacings of zirconium phos-
phonates 1–3 (19.9–21.9 nm) determined by XRPD, it can be
deduced that each filiform structure consists of stacks of 10–
15 layers of zirconium phosphonates, resulting in few mesopores
with the average dimensions of 1.9–3.2 nm (see inset of Fig. 8-2).
also synthesized by the reaction of functionalized phosphoric
acids with ZrOCl2$8H2O, but they were completely insoluble in
organic solvents. Therefore, we can conclude that the number of
phenyl, amino and hydroxyl groups in the pendant organic
moieties and their cooperative effect plays an important role in
the solubility of zirconium phosphonates 1–4 in organic
solvents.
Relation between organosolubility and filiform structure. The
solubility of 1–4 in various organic solvents is related to the
filiform structure as well as the pendant organic moieties. If
duringꢁthe preparation stirring was continued for over 12 h at
70–80 C, the zirconium phosphonates 1–4 became insoluble in
the organic solvents mentioned above, despite having the same
chemical composition. TEM of the modified zirconium phos-
phonates 1–4 showed that surprisingly the filiform morphology
of zirconium phosphonates 1–4 had been converted to the
typical amorphous (mesoporous) and crystalline (layered)
nanoparticles, with the same interlayer spacings (Fig. 9) as the
reported insoluble zirconium phosphonates. The conversion of
the filiform structure to the amorphous and crystalline states
may therefore be the reason for the organosoluble zirconium
phosphonates 1–4 becoming insoluble in organic solvents at
stirring times above 12 h.
Solubility in organic solvents
Relation between the organosolubility and the substituted group
in zirconium phosphonates. Reported zirconium phosphonates
have seldom been soluble in organic solvents; there is just one
report of an organosoluble zirconium phosphonate, with com-
position [(iPrO)3Zr(m-OiPr)2(m-OPOPh2)Zr(OiPr2]Ph2P(O)(OH),
which was prepared in high yield by the reaction of Ph2P(O)(OH)
with zirconium isopropoxide at ꢀ60 C in toluene.14 However,
ꢁ
due to the lack of complexing groups for catalytic transition
metals, this type of organosoluble zirconium phosphonate
has not received much attention in the field of catalysis. It is
therefore noteworthy that the zirconium phosphonates 1–4
synthesised in this work show good solubility (0.3–0.7 g mLꢀ1) in
various organic solvents (Table 2).
Recycling from organic solvents
We considered that the different solubilities of zirconium phos-
phonates 1–4 in organic solvents could be beneficial in achieving
homogeneous catalysis in various solvents and recycling by
From Table 2, it can be seen that 1 is readily soluble in CHCl3
and THF, with a solubility of about 0.57 and 0.59 g mLꢀ1
respectively. Zirconium phosphate 2 (where R0 is PhCH2 instead
of Ph) can be dissolved in other solvents such as toluene, benzene
and ethyl acetate (0.21–0.71 g mLꢀ1), whereas 3 and 4 (where R00
is a methyl group instead of a hydroxymethyl group) are only
soluble in THF and toluene. All four zirconium phosphonates
are insoluble in ethanol, hexane, petroleum ether and cyclo-
hexane. It is therefore clear that the solubility properties of 1–4
are related to the identity of the substituted groups (R0 and R00).
To further illustrate the relationship between the organo-
solubility and the number of phenyl groups attached to the
backbone of the zirconium phosphonates, one of the phenyl
groups at the hydroxyl group position in 1 was changed to be
a hydrogen atom instead of a phenyl group. This zirconium
phosphonate was immiscible in organic solvents,6d although
the corresponding titanium phosphonate was (slightly) soluble
only in ethanol (30–50 mg mLꢀ1).15 Zirconium phosphonates
with more phenyl groups and without amino or hydroxyl
Fig. 9 The conversion of the filiform structure of zirconium phospho-
nates to the mesoporous and layered states.
groups, Zr(OH)1.8[O3P(CH2)nPPh2]1.1 (n
Zr(OH)2.2[O3P(CH2)nN(CH2CH2PPh2)2]0.9 (n
¼
2–6) and
2–6), were
¼
Table 2 The organosolubilities of zirconium phosphonates 1–4 in
organic solvents (g mLꢀ1
a
)
Phosphonate Chloroform THF Toluene Benzene Ethyl acetate
1
2
3
4
0.57
0.46
0
0.59
0
0
0.21
0
0
0.51
0
0.65 0.71
0.47 0.57
0.52 0.53
0
0
0
a
Determined in 1 mL of solvent at 25 ꢁC.
Fig. 10 The recycling procedure of zirconium phosphonates 1–4.
J. Mater. Chem., 2009, 19, 1098–1104 | 1103
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Acknowledgements
We are grateful to the Chongqing Scientific Foundation, P. R.
China (CSTC, 2007BB4368), for the support of this work.
References
1 (a) C. E. Song, Annu. Rep. Prog. Chem., Sect. C: Phys. Chem., 2005,
101, 143; (b) A. Cormaa and H. Garciaa, Adv. Synth. Catal., 2006,
348, 1391.
2 (a) R. S. Drago and D. C. Pribich, Inorg. Chem., 1985, 24, 1983; (b)
C. E. Song, J. S. Lim and S. C. Kim, Chem. Commun., 2000, 2415;
(c) M. W. McKittrick and C. W. Jones, J. Am. Chem. Soc., 2004,
126, 3052; (d) M. W. McKittrick and C. W. Jones, J. Catal., 2004,
227, 186; (e) M. W. McKittrick and C. W. Jones, Chem. Mater.,
2003, 15, 1132.
Fig. 11 (a) UV adsorption in solution and (b) percent recovery of
zirconium phosphonate 2 with different volume ratios of good solvent to
poor solvent (v1/v2).
3 (a) C. Li, Catal. Rev., 2004, 46, 419; (b) H. Zhang, Y. Zhang and C. Li,
J. Catal., 2006, 238, 369.
means of precipitation. To assess the ease of recycling on a small
scale, the filiform zirconium phosphonates 1–4 were dissolved in
one of the good solvents (v1), which are generally suitable for
homogeneous catalytic reactions. Upon adding a poor solvent
(v2), zirconium phosphonates 1–4 precipitated out, providing
a means of separation (Fig. 10).
4 (a) V. Ayala, A. Corma and M. Iglesias, J. Catal., 2004, 224, 170; (b)
B. Pugin, H. Landert, F. Spindler and H. U. Blaser, Adv. Synth.
Catal., 2002, 344, 974; (c) Q. H. Fan, Y. M. Chen and X. M. Chen,
J. Am. Chem. Soc., 1999, 121, 7407; (d) D. Z. Jiang, F. Xi and
A. S. C. Chan, Chem. Commun., 2000, 789; (e) S. J. Liu,
H. Y. Cheng, F. Y. Zhao, J. Y. Gong and S. H. Yu, Chem. Eur. J.,
2008, 14(13), 4076; (f) J. Wang, J. C. Groen, W. Yue, W. Zhou and
M. O. Coppens, J. Mater. Chem., 2008, 18(4), 468.
For example, 0.6 mg of phosphonate 2 was dissolved in 1 mL
of the first solvent (v1), and the second solvent (v2) was then
added in a volume ratio (v2/v1) between 10:1 and 1:1. The
phosphonate 2 precipitated as a solid and was separated from the
mixture centrifugally. From the concentration of zirconium
phosphonate in the organic phase as determined by UV spec-
troscopy, the percent recovery (2–89%) could be calculated
(Fig. 11). From Fig. 11, it can be seen that THF as the first
solvent (v1) and petroleum ether as the second solvent (v2) is the
optimal system for precipitation of zirconium phosphonate 2.
Under these conditions, we found that when 50 mg of zirconium
phosphonate 1–4 was used, excellent recoveries of 95–100%
could be achieved.
5 (a) D. E. Bergbreiter, Chem. Rev., 2002, 102, 3345; (b) S. Bastin,
R. J. Eaves and C. W. Edwards, J. Org. Chem., 2004, 69, 5405; (c)
G. J. Deng, B. Yi and Y. Y. Huang, Adv. Synth. Catal., 2004, 346,
1440; (d) G. E. Oosterom, J. N. H. Reek and P. C. J. Kamer,
Angew. Chem. Int. Ed., 2001, 40, 1828; (e) R. van Heerbeek,
P. C. J. Kamer and P. W. N. M. van Leeuwen, Chem. Rev., 2002,
102, 3717.
6 (a) For examples see: A. Hu, H. L. Ngo and W. Lin, J. Am. Chem.
Soc., 2003, 125, 11490; (b) C. V. Kumar and A. Chaudari, J. Am.
Chem. Soc., 2000, 122, 830; (c) U. Costantino, M. Curini and
O. Rosati, Curr. Org. Chem., 2004, 8, 591; (d) X. Wu, X. Ma and
Y. Ji, J. Mol. Catal. A: Chemical, 2007, 265(1–2), 316; (e) B. Zheng,
X. Wu, X. Ma, L. Xiao, Y. Ji and X. Fu, Catal. Commun., 2007, 8,
1923; (f) H. L. Ngo, A. Hu and W. Lin, J. Mol. Catal. A: Chemica.,
2004, 215, 177.
7 For examples see: (a) K. Huang, J. Yu and G. Wang, Eur. J. Inorg.
Chem., 2004, 14, 2956; (b) M. J. de la Mata and O. Juanes, Angew.
Chem. Int. Ed., 2004, 43, 619; (c) J. A. Fry, C. R. Samanamu,
J. L. MontcharnpandA. F. Richards, Eur. J. Inorg. Chem., 2008, 3, 463.
8 For a review see: D. E. Bergbreiter, Topics In Current Chemistry,
2004, 242, 113.
9 R. Dragone, P. Galli, M. A. Massucci and M. Trombetta, J. Mater.
Chem., 2003, 13, 834.
10 (a) A. Clearfield and D. G. Smith, Inorg. Chem., 1969, 8, 431; (b)
G. Alberti, U. Costantino, S. Allulli and N. Tomassini, J. Inorg.
Nucl. Chem., 1978, 40, 1113.
11 J. C. Amicangelo and W. R. Leenstra, Inorg. Chem., 2005, 44(6), 2067.
12 (a) D. M. Poojary, H. L. Hu, F. L. Campbell and A. Clearfield, Acta
Crystallogr., Sect. B, 1993, 49, 996; (b) W. R. Leenstra and
J. C. Amicangelo, Inorg. Chem., 1998, 37, 5317.
13 (a) P. L. Gentili, U. Costantino, M. Nocchetti, C. Miliania and
G. Favaro, J. Mater. Chem., 2002, 12, 2872; (b) R. Vivani,
G. Alberti, F. Costantino and M. Nocchett, Microporous
Mesoporous Mater., 2008, 107, 58; (c) A. Clearfield, Journal of
Alloys and Compounds, 2006, 418, 128; (d) F. Bellezza, A. Cipiciani,
U. Costantino and S. Nicolis, Langmuir, 2004, 20, 5019.
14 D. Chakraborty, V. Chandrasekhar and M. Bhattacharjee, Inorg.
Chem., 2000, 39, 23.
Conclusions
In conclusion, we have prepared for the first time mesoporous
layered zirconium phosphonates with complexing groups and
good organosolubility. These zirconium phosphonates possess
different solubilities in organic solvents, and this can be utilized
to recycle them by precipitation. Owing to their layered meso-
porous structures, these zirconium phosphonates could act as
catalyst supports, enabling one to take advantage of the site
isolation effect, the confinement effect and the cooperation
effect, and bridging the gap between heterogeneous catalysts and
homogeneous catalysts.
Finally, if a chiral b-aminoalcohol is used in the organic
moieties of 1–4, these zirconium phosphonates could have
potential application in asymmetric transformations, such as
aldol, Mannich and Michael reactions, paving the way for the
use of solid base in homogeneous asymmetric catalysis.
15 Y. Ji, X. Ma and X. Wu, Appl. Catal. A: General, 2007, 332, 247.
1104 | J. Mater. Chem., 2009, 19, 1098–1104
This journal is ª The Royal Society of Chemistry 2009