Titanium dioxide nanoparticles functionalized with Pd and W complexes of a catecholphosphane ligand

Article abstract of DOI:10.1002/ejic.200400584

Palladium and tungsten complexes containing the bifunctional ligand 4-diphenylphosphanylcatechol (L¹) have been synthesized. The catecholate functionality strongly binds to titanium dioxide nanoparticles, effectively anchoring the complexes to

Full text of DOI:10.1002/ejic.200400584

FULL PAPER
Titanium Dioxide Nanoparticles Functionalized with Pd and W Complexes of a
Catecholphosphane Ligand
Nigel T. Lucas,[a] James M. Hook,[a] Andrew M. McDonagh,*[a,b] and
Stephen B. Colbran*[a]
Keywords: Palladium / Tungsten / Phosphane ligands / Titanium
Palladium and tungsten complexes containing the bifunc-
surface-bound tungsten carbonyl complexes. Preliminary ex-
tional ligand 4-diphenylphosphanylcatechol (L1) have been periments show that the palladium phosphane complex
synthesized. The catecholate functionality strongly binds to
titanium dioxide nanoparticles, effectively anchoring the
complexes to the TiO2 surface. Solid-state 31P NMR spectro-
scopy was used to probe the surface-bound compounds. Dif-
[PdBr2(L1)2] supported on TiO2 catalyzes the Sonogashira
coupling of phenylacetylene with 4-iodonitrobenzene.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
fuse reflectance infrared spectroscopy also provided data on Germany, 2005)
Introduction
bind metal centers as catecholate, semiquinone and quino-
noid forms.[16Ϫ18] Attractively, use of a catechol anchor
could potentially increase communication between a sur-
face-immobilized transition metal species and the titania
surface.[19] We chose phosphanes as the metal-binding
groups because phosphanes are amongst the most common
ligands in transition metal chemistry and phosphane com-
plexes of transition metals exhibit a diverse range of desir-
able photophysical and redox properties in addition to be-
ing the largest class of molecular catalysts.[5,6]
Immobilizing materials on metal oxide supports is at-
tracting sustained interest with applications in areas such
as catalysis, electrocatalysis, electrochromic display, photod-
egradation of organic compounds, and photovoltaic power
generation.[1Ϫ4] Titanium dioxide is a useful support be-
cause it is inexpensive, inert, and nontoxic, as well as having
an electronic band structure suitable for the aforemen-
tioned applications.
Scheme 1 outlines two approaches toward our goal of
phosphane-substituted metal complexes on titania. We an-
ticipated that either TiO2 particles could be functionalized
with a phosphanyl-catechol and then treated with transition
metal sources or that metal phosphane complexes bearing
the free catechol group could be prepared which would di-
rectly attach to the TiO2 surface. Phosphanes bearing phos-
phonite or alkoxysilane-substituted tethers have been pre-
viously employed to immobilize metal complexes on
SiO2,[20Ϫ32] but to the best of our knowledge not to titania,
and never before has the "non-innocent" catecholate group
been employed to anchor a transition metal phosphane
complex to a metal oxide surface. Raymond et al.[12] re-
cently reported a catecholphosphane (L1, see below) that
suited our purposes. In this paper, we present the results of
investigations on L1 and transition metal complexes of L1
immobilized on nanocrystalline TiO2 particles.
Our aim was to use a bifunctional catechol (1,2-
dihydroxybenzene)Ϫphosphane compound to anchor tran-
sition metal complexes ligated via the P atom to TiO2
bound by the catecholate group. We envisaged that the soft
phosphane donor should preferentially coordinate to soft
metal centers,[5Ϫ8] and the catechol group, which is a source
of the hard catecholate dianion, should preferentially bind
to hard TiIV metal centers.[9Ϫ18] Anchoring groups such as
carboxylate and phosphonate are regularly used to immo-
bilize molecules on TiO2 and other metal oxide surfaces.
Although it is known that catechol forms a strongly immo-
bilized complex with TiO2 surfaces,[13,14] the use of catechol
as an anchoring substituent for metal complexes has been
scarcely documented.[15] The catecholate binding group is
"non-innocent" in the sense that it is redox active, able to
[a]
School of Chemical Sciences, The University of New South
Wales,
Sydney, NSW 2052, Australia
Fax: (internat.) ϩ 61-2-9385-6141
E-mail: s.colbran@unsw.edu.au
Results and Discussion
[b]
Syntheses
Current address: Department of Chemistry, Materials and
Forensics, University of Technology Sydney,
Sydney, NSW 2007, Australia
E-mail: andrew.mcdonagh@uts.edu.au
The precursor 4-(diphenylphosphanyl)veratrole (L2) was
prepared using an adaptation of the method of Schmid et
496
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200400584
Eur. J. Inorg. Chem. 2005, 496Ϫ503

Functionalized Titanium Dioxide Nanoparticles
FULL PAPER
Scheme 1. Approaches to using a bifunctional catecholphosphane compound to anchor transition metals complexes to TiO2 surfaces
Scheme 2. Preparation of derivatives of catecholphosphane L1
al.[33] Lithiation of 4-bromoveratrole in tetrahydrofuran
The reaction of [PdCl2(PhCN)2] with L2 gave
using n-butyllithium and subsequent reaction with chloro- [PdCl2(L2)2] in 80% yield (Scheme 3). Treatment of this
diphenylphosphane gave L2 in 95% yield (Scheme 2). At- complex with boron tribromide did not yield the desired
tempted demethylation of L2 to afford L1 using boron trib- catecholphosphane complex, [PdBr2(L1)2] (1), but the bro-
romide yielded instead the boron tribromide adduct mide-bridged bi-metallic complex [Pd2Br4(L1)2], which was
L1·BBr3. The 31P NMR spectrum of L1·BBr3 has four peaks isolated in 70% yield. The H and 31P NMR spectra, ESI-
1
of equal intensity because of 31P-11B scalar coupling (11B: MS, and elemental microanalysis are all consistent with the
I ϭ 3/2, 80.4% abundance). The chemical shift, δ ϭ dimeric structure for [Pd2Br4(L1)2]. The presence of the ad-
Ϫ2.4 ppm, is consistent with other P(Ar)3BBr3 adducts.[34] duct L1·BBr3 in the crude reaction mixture (about 1:1 by
The PϪB bond was found to be quite stable and we were 1H NMR spectroscopy) suggests that one of the phosphane
unable to cleanly displace the boron tribromide by reaction groups is captured by boron tribromide resulting in a coor-
with strong Lewis bases such as di- and triethylamine, or dinatively unsaturated palladium species. Combination of
by hydrolysis with aqueous hydroxide. Demethylation of L2 two such species would give the stable bromide-bridged di-
was achieved by the procedure of Raymond et al.[12] to yield mer. A similar reaction has been previously reported by
the catecholphosphane hydrobromide, L1·HBr. Oxidation Sembiring et al.[35] Complex 1 was eventually and con-
of L2 with hydrogen peroxide followed by demethylation veniently obtained in 70% yield from the reaction of
with boron tribromide afforded the catecholphosphane ox- [PdCl2(PhCN)2] with L1·HBr followed by treatment with
ide L1ϭO in 81% overall yield.
excess KBr. During this reaction, bromide for chloride li-
Eur. J. Inorg. Chem. 2005, 496Ϫ503
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
497

N. T. Lucas, J. M. Hook, A. M. McDonagh, S. B. Colbran
FULL PAPER
Scheme 3. Preparation of catecholphosphane palladium(ii) complexes
gand metathesis occurs and stirring the crude reaction mix-
The CP/MAS 31P NMR spectrum of L1·HBr (Figure 1,
ture with excess potassium bromide ensures that the a) shows an isotropic signal at δP ϭ Ϫ3.6 ppm, (linewidth,
amount of chloride-containing complex in the product is ν1/2 ϭ 450 Hz, chemical shift anisotropy, ∆σ ϭ 100 ppm)
negligibly small.
little shifted from the solution value, δP ϭ Ϫ6.73 ppm
The tungsten(0) carbonyl complex [W(CO)5(L2)] was pre- ([D7]dimethylformamide).[12] In contrast, the CP/MAS 31P
pared by the in situ reaction of L2 with freshly prepared NMR spectrum of the phosphane oxide L1ϭO (Figure 1,
[W(CO)5(THF)] (THF ϭ tetrahydrofuran) (Scheme 4). Bo- e) yields two values for the isotropic chemical shift, δP
ron tribromide effected cleavage of the methyl groups in
ϭ
[W(CO)5(L2)], without significant decomplexation of the
phosphane from the tungsten, to afford the tungsten(0) car-
bonyl derivative [W(CO)5(L1)] (2) in 72% yield.
Scheme 4. Preparation of catecholphosphane tungsten(0) com-
plexes
Preparation of TiO2-Bound Catecholphosphane Compounds
The TiO2-bound compounds were prepared by stirring a
solution of the compound with TiO2 nanoparticles (De-
gussa P-25). The particles were then filtered and rinsed suc-
cessively with a number of solvents to ensure that no free
compound remained before drying in vacuo at room tem-
perature.
Solid-State 31P NMR Studies of TiO2-Bound
Catecholphosphane Compounds
Solid state 31P{1H} NMR spectra have been acquired of
the catecholphosphane L1, its oxide L1ϭO, and its metal
complexes, 1 and 2, adsorbed onto titanium dioxide nano-
particles as well as in their pure form. Spectra were ob-
tained using cross-polarization to enhance the NMR sensi-
tivity of the dilute species bound to titania.
Figure 1. 31P CP/MAS solid-state NMR spectra of (a) pure
L1·HBr; (b) TiO2-bound L1; (c) pure 2; (d) TiO2-bound 2; (e) pure
L1ϭO; (f) TiO2-bound L1ϭO; for all spectra, the temperature was
303 K and the spin rate was 5 kHz, except for (e): 3 kHz
498
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 496Ϫ503

Functionalized Titanium Dioxide Nanoparticles
FULL PAPER
36.0 and 33.9 ppm, suggestive of two chemically distinct
phosphorus environments either within the crystal lattice or
in two crystal polymorphs. The chemical shift anisotropy,
∆σ ϭ 200 ppm, is indicative of a less isotropic electronic
distribution about the phosphorus(v) atom in L1ϭO than
about the phosphorus(iii) atom in L1·HBr. The magnitude
of the chemical shift anisotropy for L1ϭO is identical to
the literature value for OPPh3.[20] Interestingly, the line-
widths for L1ϭO, ν1/2 ϭ 145 Hz, are appreciably narrower
than those for L1·HBr, perhaps reflecting increased crystal-
linity for the L1ϭO sample.
It was anticipated that immobilization of L1 on the TiO2
surface would provide titania particles functionalized with
phosphane groups. The spectra of L1·HBr adsorbed to TiO2
(Figure 1, b) yielded an isotropic chemical shift, δP
ϭ
Ϫ2.8 ppm, similar to that for unbound L1·HBr. The spectra
show no peaks for the phosphane oxide L1ϭO (see below).
Clearly, the catechol group effectively immobilizes L1 on the
TiO2 surface without significant oxidation of the phos-
phane functionality. The linewidth of L1 on TiO2, ν1/2
ϭ
750 Hz, is of a comparable order of magnitude to that for
pure L1·HBr, ν1/2 ϭ 400 Hz. Likewise, the CP/MAS 31P
NMR spectrum of pure 2 shows an isotropic signal at δP ϭ
20.1 ppm (Figure 1, c), while that of TiO2-bound 2 (Fig-
ure 1, d) is at δP ϭ 19.9 ppm. Again the linewidths of 2,
ν1/2 ϭ 370 Hz, and TiO2-bound 2, ν1/2 ϭ 260 Hz, are com-
parable. The relatively invariant chemical shift dispersion
upon immobilization of ligand L1 or complex 2 is indicative
of the surface-bound species, in each case, occupying near
identical environments on the TiO2 surface. In each case,
the surface sites occupied by the immobilized species ap-
pear remarkably the same.
The CP/MAS 31P NMR spectrum of phosphane oxide
L1ϭO adsorbed to TiO2 (Figure 1, f) yields an isotropic
chemical shift, δP ϭ 39.7 ppm, and chemical shift aniso-
tropy, ∆σ ϭ 200 ppm, suggesting the local environment of
the phosphorus atom does not appreciably change upon
binding to TiO2. The linewidth for TiO2-bound L1ϭO,
Figure 2. (a) FTIR spectrum of 2 in dichloromethane solution; (b)
diffuse reflectance FTIR spectrum of solid 2; (c) diffuse reflectance
FTIR spectrum of TiO2-supported 2
ν
1/2 ϭ 1.5 kHz, however, is substantially broader than an-
ticipated. Phosphane oxides are known to bind to TiO2
surfaces,[36Ϫ38] and it is possible that, for TiO2-bound L1ϭ
O, the phosphane oxide oxygen atom also binds to the sur-
face, which leads to disordered surface environments and
The diffuse reflectance infrared spectrum of a solid
consequently to substantial chemical shift dispersion. This sample of 2 is shown in Figure 2, b. The A1eq band appears
is in agreement with the trend observed for solid PPh3 and at 2071 cmϪ1, a value similar to that observed in the solu-
OPPh3 bound to silica surfaces.[20Ϫ23,25Ϫ30,32,36Ϫ38]
tion spectrum. Most noticeable is the splitting of the E in-
frared modes, with peaks at 1967, 1926, and 1900 cmϪ1
,
due to different carbonyl environments created by packing
in the solid state.
Infrared Experiments on TiO2-Bound Tungsten Complexes
The strong infrared-active carbonyl absorption bands of
The diffuse reflectance infrared spectrum of TiO2-bound
the tungsten(0) carbonyl 2 make it an appropriate probe for 2 is shown in Figure 2, c. The A1eq band is seen at 2073
the binding of the catechol group of L1 to titania. Figure 2 cmϪ1, almost unchanged from the previous two spectra. A
shows solution and diffuse reflectance infrared spectra of 2. broad E band with a transmittance minimum at approxi-
The solution spectrum of 2 (Figure 2, a) is characteristic of mately 1956 cmϪ1 is observed with a much less symmetrical
complexes of the type [W(CO)5(PAr3)]. The bands at ap- shape than that in the solution spectrum. This may also be
proximately 2070 and 1935 cmϪ1 are assigned to the A1eq attributed to packing-induced anisotropy and indicates the
and E vibrational modes,[39] respectively, and are associated TiO2-bound 2 also experiences different carbonyl environ-
with the four equivalent carbonyl ligands.
ments, created by packing and orientation on the surface.
Eur. J. Inorg. Chem. 2005, 496Ϫ503
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
499

N. T. Lucas, J. M. Hook, A. M. McDonagh, S. B. Colbran
FULL PAPER
To explore if TiO2 particles treated with L1·HBr react version of 4-iodonitrobenzene to (4-nitrophenyl)phenyle-
with a metal center, we treated [W(CO)5(THF)] with TiO2 thyne). The reaction mixture was decanted from the settled
particles bearing surface-bound ligands. The diffuse reflec- catalyst, and fresh solvent, reactants, and copper(i) iodide
tance infrared spectrum was recorded. Bands at 2072, 1984 were added. After the same reaction time and conditions as
and a broad strong band at approximately 1955 cmϪ1 may the first batch, the reaction had again proceeded to com-
1
be
attributed
to
successful
reaction
between pletion (as evidenced by H NMR spectroscopy). The pro-
[W(CO)5(THF)] and L1-derivatized TiO2 to afford TiO2- cess was repeated a third time whereupon the reaction re-
bound 2. We also found that stirring [W(CO)5(THF)] with ached about 75% completion after the same reaction time,
untreated TiO2 particles followed by washing resulted in a because of the deactivation of the palladium catalyst. A
surface-bound tungsten carbonyl species. The diffuse reflec- control reaction was also performed where unmodified
tance infrared spectrum of this species exhibits character- TiO2 was added to a triethylamine solution of 4-iodonitrob-
istic carbonyl stretching bands. Comparison with solution enzene and phenylacetylene, together with a cocatalyst, cop-
[40,41]
infrared spectra of [W(CO)5OH]Ϫ
gives some indi- per(i) iodide. After 24 h, no coupling was detected using 1H
cation of the nature of the bound tungsten carbonyl species. NMR spectroscopy.
These spectra are similar to the surface-bound tungsten
These preliminary results suggest that palladium com-
carbonyl species formed upon reaction of [W(CO)5(THF)] plexes anchored to TiO2 by catecholphosphane ligands do
with TiO2. It appears that the species formed involves a function as catalysts, although with limited turnover cycles.
TiϪOϪW bond such as that illustrated in Figure 3.
A full study is required to establish the cause of deacti-
vation of the palladium catalyst as well as comparative
studies with existing catalysts. Overall, the amount of pal-
ladium employed is no larger than that used in conventional
homogeneous catalysis (typically 1Ϫ5 mol %). Because of
the ease of catalyst separation (simple filtering from the re-
action mixture) this TiO2-confined catalyst may be useful
in situations where product and catalyst are difficult to sep-
arate such as where the reagents and products have highly
polar groups and are difficult to purify by the usual chrom-
atographic procedures.
Figure 3. A possible surface-bound tungsten carbonyl species re-
sulting from the reaction of [W(CO)5(THF)] and TiO2
Coupling of Phenylacetylene with 4-Iodonitrobenzene
Catalyzed by [PdBr2(L1)2] Supported on TiO2
Conclusion
Palladium complexes are effective catalysts in the coup-
ling reaction between aryl halides and terminal alkynes
(Sonogashira coupling).[42Ϫ43,44,45] In order to establish the
potential of surface-immobilized catecholphosphane metal
complexes in catalysis, a preliminary study of the coupling
between 4-iodonitrobenzene and phenylacetylene in the
presence of the palladium complex 1 supported on TiO2
(Scheme 5) was undertaken.
Our results reveal that TiO2-immobilized transition
metalϪphosphane complexes are readily accessible starting
from catecholphosphanes. Infrared experiments indicate
binding of a pre-formed catecholphosphane complex to
TiO2 to give a more precisely defined material than does
sequential surface derivatization with the catecholphos-
phane ligand and subsequent reaction with a metal precur-
sor. Further investigations of applications, such as those
summarized in the Introduction, are clearly warranted.
Experimental Section
Scheme 5. Coupling reaction between 4-iodonitrobenzene and phe-
nylacetylene catalyzed by CuI and TiO2-supported 1
General Remarks: Reactions were carried out under nitrogen by
using Schlenk techniques. All solvents were analytical grade (AR).
The following reaction solvents were dried and distilled under ni-
The supported catalyst was prepared by stirring a di-
chloromethane solution of 1 with TiO2 at room temperature trogen using standard methods: dichloromethane over calcium hy-
dride; tetrahydrofuran over sodium or potassium benzophenone
ketyl. Acetone, diethyl ether, absolute ethanol and triethylamine
were used without drying but were purged with nitrogen prior to
use. Column chromatography was performed using Merck silica
gel 60 of particle size 0.040Ϫ0.063 mm (230Ϫ400 mesh ASTM).
Analytical TLC was carried out on Merck aluminum sheets coated
for 24 h (no special treatment was undertaken to remove
any TiO2 surface-bound water). The suspension was filtered
and rinsed successively with dichloromethane, acetone,
methanol and diethyl ether to ensure that no unbound ma-
terial remained. The estimated loading of 4 on titania was
1Ϫ2% w/w. The TiO2-supported 1 was added to a triethyl-
amine solution of 4-iodonitrobenzene and phenylacetylene,
together with a cocatalyst, copper(i) iodide. The reaction
proceeded to completion after stirring at room temperature
with 0.25 mm silica gel 50 PF254
.
The reagents 4-bromoveratrole, n-butyllithium, boron tribromide,
tungsten hexacarbonyl, phenylacetylene (Aldrich), hydrogen per-
oxide (20%), MnO2, KBr, 4-nitroiodobenzene and CuI were pur-
for 24 h (1H NMR spectroscopy showed quantitative con- chased commercially and used as received. Chlorodiphenylphos-
500
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 496Ϫ503

Functionalized Titanium Dioxide Nanoparticles
FULL PAPER
1
2
phane (Aldrich) was distilled before use. The titanium dioxide
(TiO2) was Degussa P25 (ca. 25-nm particles). Literature pro-
cedures were used to prepare 4-(diphenylphosphanyl)catechol
hydrobromide (L1·HBr),[12] and [PdCl2(NCPh)2].[46]
2JH,H ϭ 2 Hz, 1 H, 5-C6H3), 7.38 (dd, JPH ϭ 12, JH,H ϭ 2 Hz, 1
H, 3-C6H3), 8.70 (br. s, 1 H, OH), 7.95Ϫ7.57 (m, 10 H, Ph), 8.92
(br. s, 1 H, OH) ppm. 31P NMR (CDCl3): δ ϭ Ϫ2.4 (1:1:1:1 q,
JBP ϭ 149 Hz).
1H and 31P NMR solution spectra were recorded in CDCl3 (Cam-
bridge Isotope Laboratories or Aldrich) or [D6]acetone (Aldrich)
using a Bruker Avance 300 spectrometer (at 300 MHz for 1H,
121 MHz for 31P). The 1H NMR spectra were referenced internally
to residual protic solvent, and 31P to external 85% H3PO4.
Solid state 31P NMR spectra were obtained using a Varian Inova
300 NMR spectrometer at 121.395 MHz using Chemagnetics cross-
polarization magic-angle spinning (CP/MAS) probes. The samples
were packed into 4 mm or 7.5 mm o.d. zirconia rotors and spun at
between 3 and 5 kHz. Spectra were recorded at 303 K and refer-
enced to the CP/MAS spectrum of external solid ammonium dihy-
drogen phosphate, δP ϭ 1.0 ppm.[47] Cross-polarized spectra were
obtained after optimizing by using the following parameters: recy-
cle time, 20Ϫ120 s; contact time, 4Ϫ5 ms; 1H 90° pulse width,
5Ϫ6 µs.
Synthesis of L2؍O: A solution of L2 (1.02 g, 3.2 mmol) in acetone
(50 mL) was cooled to 0 °C and hydrogen peroxide solution
(3.5 mL) was added dropwise with stirring. The mixture was stirred
at 0 °C for 2 h at which point TLC indicated that the reaction was
complete. MnO2 was added in batches until no effervescence was
observed, then the suspension was stirred for a further 1 h. The
brown residue was removed by paper filtration, the filtrate tested
for peroxides before being taken to dryness in a rotary evaporator
to yield a cream colored solid identified as L2ϭO (0.90 g, 84%).10,11
1H NMR (CDCl3): δ ϭ 3.83 (s, 3 H, Me), 3.90 (s, 3 H, Me), 6.87
1
2
1
(dd, JH,H ϭ 8, JPH ϭ 3 Hz, 1 H, 6-C6H3), 7.01 (ddd, JPH ϭ 12,
2
1
1JH,H ϭ 8, JH,H ϭ 2 Hz, 1 H, 5-C6H3), 7.28 (dd, JPH ϭ 12,
2JH,H ϭ 2 Hz, 1 H, 3-C6H3), 7.39Ϫ7.54 (m, 6 H, Ph), 7.59Ϫ7.67
(m, 4 H, Ph) ppm. 31P NMR (CDCl3): δ ϭ 32.0 ppm.
Infrared spectra were recorded on a Nicolet Avatar 320 FT-IR
spectrometer in a solution cell or using a diffuse reflectance infra-
red FT (DRIFTS) attachment. The background for DRIFTS
samples was air (for neat samples) or oven-dried TiO2 (for sup-
ported samples). Electrospray ionization mass spectra (ESI-MS)
were acquired on a VG Quattro mass spectrometer with a capillary
voltage of 4 kV and a cone voltage of 30 V. Elemental microanaly-
ses were carried out by the Microanalytical Service Unit at the
Research School of Chemistry, Australian National University.
Synthesis of L1؍O: A solution of L2ϭO (900 mg, 2.7 mmol) in
dichloromethane (80 mL) was cooled to Ϫ78 °C and BBr3
(0.80 mL, 8.4 mmol) was added dropwise with stirring. The mixture
was allowed to warm to room temperature over 18 h. Dilute NaOH
solution was carefully added to destroy excess BBr3, then the reac-
tion mixture was transferred to a separating funnel, acidified with
HCl (5 m), and the dichloromethane layer was collected. The aque-
ous phase was extracted again with dichloromethane (2 ϫ 25 mL).
The dichloromethane washings were combined and washed with
dilute acid (ca. 0.1 m HCl, 20 mL), dried with MgSO4, filtered, and
taken to dryness in a rotary evaporator. A cream-colored powder
was isolated and identified as the hydrate of the product, L1ϭ
Synthesis of 4-(Diphenylphosphanyl)veratrole (L2): A solution of 4-
bromoveratrole (4.00 mL, 27.8 mmol) in freshly distilled tetra-
hydrofuran (110 mL) was cooled to Ϫ78 °C and n-butyllithium
solution (1.55 m) was added dropwise over 3 min. After stirring for
15 min, chlorodiphenylphosphane was added dropwise to the co-
oled solution. The stirred solution was allowed to warm to room
temperature over 17 h after which all volatile materials were re-
moved in vacuo. The white residue was extracted several times with
dichloromethane and the washings taken to dryness in a rotary
evaporator. The crude material was loaded onto a 15 ϫ 3 cm silica
column and eluted with petroleum spirit/dichloromethane (up to
1:1) with the collection of fractions. Fractions containing the prod-
uct (as identified by analytical TLC, Rf ϭ 0.55, petroleum ether/
dichloromethane, 1:1) were combined and taken to dryness using a
rotary evaporator to afford whiteϪpale green crystals identified as
4-(diphenylphosphanyl)veratrole (L2) (8.50 g, 26.4 mmol, 95%).[5,11]
1H NMR (CDCl3): δ ϭ 3.76 (s, 3 H, Me), 3.92 (s, 3 H, Me),
6.92Ϫ6.86 (m, 3 H, C6H3), 7.37Ϫ7.28 (m, 10 H, Ph) ppm. 31P
NMR (CDCl3): δ ϭ Ϫ3.2 ppm.
1
1
O·H2O (800 mg, 96%). H NMR (CDCl3): δ ϭ 6.61 (ddd, JPH
ϭ
1
2
1
12, JH,H ϭ 8, JH,H ϭ 2 Hz, 1 H, 5-C6H3), 6.85 (dd, JH,H ϭ 8,
2JPH ϭ 4 Hz, 1 H, 6-C6H3), 7.40Ϫ7.65 (m, 10 H, Ph), 8.00 (br. m,
1
2
2 H, OH), 7.81 (dd, JPH ϭ 13, JH,H ϭ 2 Hz, 1 H, 3-C6H3) ppm.
31P NMR (CDCl3): δ ϭ 35.3 ppm. 31P CP/MAS NMR: δ ϭ 33.9,
36.1 (about 3:1). C18H15O3P·H2O (328.30): calcd. C 65.85, H 5.22;
found C 66.82, H 4.87.
Synthesis of [PdCl2(L2)2]: A solution of [PdCl2(NCPh)2] (500 mg,
1.30 mmol) in dichloromethane (15 mL) was added to a solution
of L2 (890 mg, 2.76 mmol) in dichloromethane (15 mL) at room
temperature. The resulting clear yellow solution was stirred for
30 min, the volume was reduced to about 10 mL, and methanol
was added dropwise to give a yellow precipitate. The mixture was
cooled in ice, and the precipitate was collected on a sintered glass
funnel, washed with methanol and dried at the pump. Slow dif-
fusion of diethyl ether into an acetone solution of the crude mate-
rial yielded yellow-orange prisms of [PdCl2(L2)2] (862 mg, 80%).
1H NMR (CDCl3): δ ϭ 6.86 (br. d, 2 H, C6H3), 3.89 (s, 6 H, Me),
3.75 (s, 6 H, Me), 7.11Ϫ7.18 (m, 2 H, C6H3), 7.33Ϫ7.45 (m, 12 H,
Ph), 7.54Ϫ7.59 (m, 2 H, C6H3), 7.71Ϫ7.63 (m, 8 H, Ph) ppm. 31P
NMR (CDCl3): δ ϭ 25.3 ppm. C40H38Cl2O4P2Pd (822.01): calcd. C
58.45, H 4.66; found C 57.55, H 4.72.
Attempted Synthesis of 4-(Diphenylphosphanyl)catechol (L1): A
solution of 4-(diphenylphosphanyl)veratrole (982 mg, 3.0 mmol) in
dichloromethane (20 mL) was cooled to Ϫ78 °C and BBr3
(1.05 mL, 11.1 mmol) was added dropwise with stirring. The mix-
ture was stirred for a further 10 min before the cooling bath was
removed and the mixture allowed to warm to room temperature
over 3 h. Methanol (1 mL) was carefully added to quench the ex-
cess BBr3, followed by water (5 mL), and the mixture was trans-
ferred to a separating funnel. The aqueous phase was neutralized
and the dichloromethane phase collected, dried with MgSO4, fil-
Synthesis of [PdBr2(L1)2] (1): A solution of [PdCl2(NCPh)2]
(240 mg, 0.63 mmol) in acetone (20 mL) was added to a stirred
suspension of L1·HBr (496 mg, 1.32 mmol) in acetone (30 mL) at
room temperature. The resulting clear orange solution was stirred
tered, and taken to dryness on a rotary evaporator. A fluffy white for a further 1 h, then KBr (6.30 g, 53 mmol) was added and the
solid was isolated and identified as the adduct, 4-
suspension stirred vigorously for 16 h. The mixture was filtered and
(diphenylphosphanyl)catecholϪboron tribromide
(L1·BBr3)
taken to dryness to yield [PdBr2(L1)2] (1) (375 mg, 70%) as a yellow
1
(1.38 g, 83%). 1H NMR ([D6]acetone): δ ϭ 7.04 (dd, JH,H ϭ 8, solid. 1H NMR ([D6]acetone): δ ϭ 6.92 (br. d, 2 H, C6H3),
1
1
2JPH ϭ 3 Hz, 1 H, 6-C6H3), 7.14 (ddd, JPH ϭ 11, JH,H ϭ 8, 7.16Ϫ7.28 (m, 4 H, C6H3), 7.37Ϫ7.45 (m, 12 H, Ph), 7.64Ϫ7.72
Eur. J. Inorg. Chem. 2005, 496Ϫ503
www.eurjic.org © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 501

N. T. Lucas, J. M. Hook, A. M. McDonagh, S. B. Colbran
FULL PAPER
(m, 8 H, Ph) ppm. 31P NMR ([D6]acetone): δ ϭ 23.7 ppm. Negative
L1 on TiO2: L1·HBr (320 mg, 0.85 mmol) was added to a rapidly
stirred suspension of TiO2 (3.4 g) in methanol (150 mL). The color
ion ESI-MS (methanol): 935 ([M ϩ Br]Ϫ, 100), 854 ([M Ϫ H]Ϫ,
55). Positive ion ESI-MS (methanol): 775 ([M Ϫ Br]ϩ, 100). of the particles changed rapidly from white to pale orange. After
C36H30Br2O4P2Pd (854.81): calcd. C 50.59, H 3.54; found C 49.64,
H 4.07.
stirring at room temperature for 1 h, triethylamine (0.5 mL,
3.6 mmol) was added, and the stirring was continued for a further
3 h. The suspension was then allowed to settle, and the colorless
solution was removed using a cannula. The material was washed
with methanol (150 mL), then filtered and rinsed successively with
methanol (150 mL), acetone (2 ϫ 50 mL), tetrahydrofuran (2 ϫ
50 mL), dichloromethane (2 ϫ 50 mL), and diethyl ether (2 ϫ
15 mL), and then dried in vacuo at room temperature for 8 h. 31P
CP/MAS NMR: δ ϭ Ϫ2.8 ppm.
Reaction of [PdCl2(L2)2] with BBr3: A solution of [PdCl2(L2)2]
(1.00 g, 1.2 mmol) in dichloromethane (30 mL) was cooled to Ϫ80
°C, and BBr3 (0.56 mL, 5.9 mmol) was added dropwise. The reac-
tion mixture was allowed to warm to room temperature overnight,
the excess BBr3 quenched with water, and the mixture was trans-
ferred to a separating funnel. Diethyl ether was added, and the
organic phase was washed with brine (acidified with HCl), dried,
filtered and taken to dryness. The crude material was dissolved in
acetone (30 mL), KBr (5.8 g, 4.9 mmol) was added, and the suspen-
sion was stirred vigorously for 16 h. The mixture was transferred
to a separating funnel, diethyl ether (30 mL) was added, and the
resulting organic layer was washed twice and then dried over
MgSO4. The mixture was filtered and taken to dryness to give an
orange solid identified as [Pd2Br4(L1)2] (375 mg, 70%). 1H NMR
([D6]acetone): δ ϭ 6.95 (dd, 1JH,H ϭ 8, 2JPH ϭ 3 Hz, 2 H, 5-C6H3),
L1؍O on TiO2: TiO2 (1.50 g) was added to a colorless solution of
L1ϭO (203 mg, 0.65 mmol) in methanol (30 mL), and the suspen-
sion was stirred at room temperature for 18 h. The pale orange
suspension was filtered and rinsed successively with methanol (3 ϫ
15 mL) and diethyl ether (10 mL), and then dried in vacuo at room
temperature for 6 h. 31P CP/MAS NMR: δ ϭ 40.1 ppm.
[W(CO)5(L1)] on TiO2: TiO2 (1.0 g) was added to a solution of
[W(CO)5(L1)] (100 mg, 0.16 mmol) in dichloromethane (100 mL),
and the suspension was stirred at room temperature for 24 h. The
pale orange suspension was then filtered and rinsed successively
with dichloromethane (5 ϫ 20 mL), acetone (2 ϫ 20 mL), and di-
ethyl ether (2 ϫ 20 mL), then dried in vacuo at room temperature
for 3 h. 31P CP/MAS NMR: δ ϭ 19.91 ppm.
1
2
7.08Ϫ7.17 (m, 2 H, 6-C6H3), 7.28 (dd, JPH ϭ 13, JH,H ϭ 2 Hz, 2
H, 3-C6H3), 7.43Ϫ7.59 (m, 12 H, Ph), 7.67Ϫ7.77 (m, 8 H, Ph), 8.67
(s, 4 H, OH) ppm. 31P NMR (CDCl3): δ ϭ 36.8 ppm. Negative ion
ESI-MS (methanol): 1121 ([M]Ϫ, 100). Positive ion ESI-MS (meth-
anol): 775 ([M/2]ϩ, 45). C36H30Br4O4P2Pd2 (1121.04): calcd. C
38.57, H 2.70; found C 38.39, H 2.68.
Synthesis of [W(CO)5(L2)]: L2 (1.38 g, 4.3 mmol) was added to a
rapidly stirred yellow solution of [W(CO)5(THF)], obtained by irra-
[W(CO)5(THF)] on TiO2: Photolytically generated [W(CO)5(THF)]
(10 mL, 28.4 mm in tetrahydrofuran, 0.284 mmol) was added to
diating [W(CO)6] (1.50 g, 4.3 mmol) in tetrahydrofuran (60 mL) TiO2 (1.0 g) and the suspension was stirred at room temperature
with a mercury lamp for 3 h (with N2 purge). The mixture was
stirred at room temperature for 2 h after which all the volatile mate-
for 24 h. The white-gray suspension was then filtered and rinsed
successively with tetrahydrofuran (3 ϫ 40 mL), acetone (2 ϫ
rials were removed in vacuo. The pale yellow residue was purified 40 mL), and diethyl ether (2 ϫ 40 mL), then dried in vacuo at room
on a silica column by eluting with petroleum spirit/dichlorometh-
ane mixtures (up to 1:1). Fractions were collected and compared by
using analytical thin-layer chromatography; the major band (pale
yellow) was evaporated to dryness to give a pale yellow solid, ident-
temperature for 3 h.
[PdBr2(L1)2] on TiO2: TiO2 (1.50 g) was added to an orange-yellow
solution of [PdBr2(L1)2] (150 mg, 0.18 mmol) in dichloromethane
(100 mL), and the suspension was stirred at room temperature for
24 h. The suspension was then filtered and rinsed successively with
dichloromethane (2 ϫ 20 mL), acetone (2 ϫ 15 mL), methanol (2
ϫ 15 mL), and diethyl ether (2 ϫ 15 mL), then dried in vacuo at
room temperature for 8 h.
1
ified as [W(CO)5(L2)] (1.69 mg, 61%). H NMR (CDCl3): δ ϭ 3.79
(s, 3 H, Me), 3.91 (s, 3 H, Me), 6.87Ϫ6.98 (m, 2 H, 5,6-C6H3), 7.07
1
(br. d, JPH ϭ 13 Hz, 1 H, 3-C6H3), 7.41Ϫ7.47 (m, 10 H, Ph);
([D6]acetone): δ ϭ 3.74 (s, 3 H, Me), 3.88 (s, 3 H, Me), 6.98Ϫ7.15
(m, 3 H, C6H3), 7.48Ϫ7.57 (m, 10 H, Ph) ppm. 31P NMR (CDCl3):
δ ϭ 23.0 (JWP ϭ 242 Hz) ppm; ([D6]acetone): δ ϭ 21.9 (JWP
242 Hz) ppm. C25H19O7PW (646.25): calcd. C 46.46, H 2.96; found
C 46.82, H 2.93.
ϭ
Catalytic Studies
Coupling of Phenylacetylene with 4-Iodonitrobenzene Catalyzed by
[PdBr2(L1)2] Supported on TiO2: Phenylacetylene (0.15 mL,
1.3 mmol), CuI (10 mg, 0.05 mmol) and [PdBr2(L1)2]-TiO2
(800 mg, ca. 1% w/w) was added to a solution of 4-iodonitroben-
zene (80 mg, 0.32 mmol) in triethylamine (30 mL, N2-purged), and
the suspension was stirred at room temperature for 24 h. The par-
ticles were allowed to settle, and the solution was transferred to a
second flask using a cannula. The solid residue was washed with
triethylamine (20 mL). The combined reaction solution and wash-
ing was taken to dryness, and the crude residue was passed through
a short silica plug by using CH2Cl2/hexane (3:7) eluent. The 1H
NMR spectrum of the resulting pale yellow solid indicated quanti-
tative conversion of 4-iodonitrobenzene to the product (4-nitrophe-
nyl)phenylethyne[48] (Ͻ1% of the homo-coupled tolane product was
also observed). A second batch of reactants and CuI catalyst (same
scale) was added to a triethylamine (30 mL) suspension of the same
Pd catalyst from above. Again, quantitative conversion into (4-
nitrophenyl)phenylethyne was observed. The cycle was repeated a
Synthesis of [W(CO)5(L1)]: A solution of [W(CO)5(L2)] (615 mg,
0.95 mmol) in dichloromethane (10 mL) was cooled to Ϫ80 °C, and
BBr3 (0.30 mL, 3.2 mmol) was added dropwise. The mixture was
allowed to warm to about Ϫ10 °C over 1.5 h, before it was cooled
again to Ϫ80 °C, and NEt3 (1 mL) and methanol (1 mL) were ad-
ded to quench the excess BBr3. After warming to room tempera-
ture, water (1 mL) was carefully added, and the mixture was trans-
ferred to a separating funnel and washed with dilute HCl solution
(3 ϫ 20 mL). The organic phase was dried over MgSO4, filtered
and taken to dryness to yield [W(CO)5(L1)] (2) (425 mg, 72%) as a
cream-colored powder. 1H NMR ([D6]acetone): δ ϭ 6.91Ϫ7.03 (m,
3 H, C6H3), 7.47Ϫ7.55 (m, 10 H, Ph), 8.41 (s, 1 H, OH), 8.53 (s, 1
H, OH) ppm. 31P NMR ([D6]acetone): δ ϭ 20.8 (JWP ϭ 241 Hz).
31P CP/MAS NMR: δ ϭ 20.1 ppm. Negative ion ESI-MS (meth-
anol): 1236 ([2M]Ϫ dimer, 100). C23H15O7PW (618.19): calcd. C
44.69, H 2.45; found C 44.98, H 2.45.
Adsorption of Phosphanylcatechol-Containing Compounds onto third time using the same Pd catalyst with a reduction of conver-
TiO2
sion to 75%.
502
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 496Ϫ503

Functionalized Titanium Dioxide Nanoparticles
FULL PAPER
[25]
[26]
[27]
[28]
[29]
O. Krocher, R. A. Koppel, M. Froba, A. Baiker, J. Catal. 1998,
178, 284Ϫ298.
P. Braunstein, H. P. Kormann, W. Meyer-Zaika, R. Pugin, G.
Schmid, Chem. Eur. J. 2000, 6, 4637Ϫ4646.
C. M. Crudden, D. Allen, M. D. Mikoluk, J. Sun, Chem. Com-
mun. 2001, 1154.
A. J. Sandee, J. N. H. Reek, P. C. J. Kamer, P. W. N. M. van
Leeuwen, J. Am. Chem. Soc. 2001, 123, 8468Ϫ8476.
H. H. Wagner, H. Hausmann, W. F. Holderich, J. Catal. 2001,
203, 150Ϫ156.
C. Merckle, J. Blumel, Chem. Mater. 2001, 13, 3617Ϫ3623.
P. W. N. M. van Leeuwen, A. J. Sandee, J. N. H. Reek, P. C. J.
Kamer, J. Mol. Catal. A-Chem. 2002, 182, 107Ϫ123.
Z. L. Lu, E. Lindner, H. A. Mayer, Chem. Rev. 2002, 102,
3543Ϫ3578.
R. Schmid, J. Foricher, M. Cereghetti, P. Schönholzer, Helv.
Chim. Acta 1991, 74, 370Ϫ389.
E. Muylle, G. P. Van Der Kelen, Spectrochim. Acta A: Mol.
Biomol. Spectr. 1976, 32, 599Ϫ603.
S. B. Sembiring, S. B. Colbran, D. C. Craig, M. L. Scudder, J.
Chem. Soc., Dalton Trans. 1995, 22, 3731Ϫ3741.
G. Guerrero, P. H. Mutin, A. Vioux, Chem. Mater. 2001, 13,
4367Ϫ4373.
B. Hu, I. D. Gay, Langmuir 1995, 11, 3845Ϫ3847.
L. Baltusis, J. S. Frye, G. E. Maciel, J. Am. Chem. Soc. 1986,
108, 7119Ϫ7120.
Acknowledgments
The authors thank Chris Blake at the University NMR Centre,
Australian National University, for preliminary solid-state NMR
measurements.
[1]
D. W. Goodman, J. Catal. 2003, 216, 213Ϫ222.
A. Fujishima, T. N. Rao, D. A. Tryk, J. Photochem. Photobiol.
[2]
C: Photochem. Rev. 2000, 1, 1Ϫ21.
[3]
[30]
[31]
M. Grätzel, Prog. Photovoltaics: Res. Appl. 2000, 8, 171Ϫ185.
[4]
D. J. Cole-Hamilton, Science 2003, 299, 1702Ϫ1706.
E. W. Abel, F. G. A. Stone, G. Wilkinson, J. A. Labinger, M.
[5]
[32]
[33]
[34]
[35]
[36]
J. Winter (Eds.), Comprehensive Organometallic Chemistry II,
Pergamon Press, Oxford, 1995.
B. Cornils, W. A. Herrman, Applied Homogeneous Catalysis
[6]
with Organometallic Compounds, VCH, Weinheim (Germany),
1996.
S. E. Watkins, D. C. Craig, S. B. Colbran, J. Chem. Soc., Dalton
[7]
Trans. 2002, 2423Ϫ2436.
S. E. Watkins, D. C. Craig, X.-H. Yang, S. B. Colbran, J. Chem.
[8]
Soc., Dalton Trans. 1999, 1539Ϫ1540.
[9]
M. Albrecht, R. J. Frohlich, J. Am. Chem. Soc. 1997, 119,
[37]
[38]
1656Ϫ1661.
[10]
M. Albrecht, Chem. Eur. J. 1997, 3, 1466Ϫ1471.
M. Albrecht, M. Napp, M. Schneider, P. Weis, R. Fröhlich,
[11]
[39]
[40]
[41]
[42]
[43]
[44]
Chem. Eur. J. 2001, 7, 3966Ϫ3975.
R. J. Angelici, M. D. Malone, Inorg. Chem. 1967, 9,
[12]
X. Sun, D. W. Johnson, D. L. Caulder, K. N. Raymond, E. H.
1731Ϫ1736.
Wong, J. Am. Chem. Soc. 2001, 123, 2752Ϫ2763.
D. J. Darensbourg, M. L. Meckfessel Jones, J. H. Reibenspies,
Inorg. Chem. 1993, 32, 4675Ϫ4676.
J. L. Cihonski, R. A. Levenson, Inorg. Chem. 1975, 14,
1717Ϫ1720.
S. Takahashi, Y. Kuroyama, K. Sonogashira, N. Hagihara,
Synthesis 1980, 627Ϫ630.
F. Quignard, S. Larbot, S. Goutodier, A. Choplin, J. Chem.
Soc., Dalton Trans. 2002, 1147Ϫ1152.
C. C. Tzschucke, C. Markert, H. Glatz, W. Bannwarth, Angew.
Chem. Int. Ed. 2002, 41, 4500Ϫ4503.
E. Gonthier, R. Breinbauer, Synlett 2003, 1049Ϫ1051.
P. Braunstein, R. Bender, J. Jud, Inorg. Synth. 1989, 26,
341Ϫ350.
[13]
´
R. Rodrıguez, M. A. Blesa, A. E. Regazzoni, J. Colloid In-
terface Sci. 1996, 177, 122Ϫ131.
[14]
[15]
P. A. Connor, K. D. Dobson, A. J. McQuillan, Langmuir 1995,
11, 4193Ϫ4195.
C. R. Rice, M. D. Ward, M. K. Nazeeruddin, M. Grätzel, New
J. Chem. 2000, 24, 651Ϫ652.
C. G. Pierpont, Coord. Chem. Rev. 2001, 216-217, 99Ϫ125.
C. G. Pierpont, C. W. Lange, Prog. Inorg. Chem. 1994, 41,
331Ϫ442.
C. G. Pierpont, R. M. Buchanan, Coord. Chem. Rev. 1981,
38, 45Ϫ87.
M. Grätzel, Pure Appl. Chem. 2001, 73, 459Ϫ467.
L. Bemi, H. C. Clark, J. A. Davies, C. A. Fyfe, R. E. Wasyl-
ishen, J. Am. Chem. Soc. 1982, 104, 438Ϫ445.
R. A. Komoroski, A. J. Magistro, P. P. Nicholas, Inorg. Chem.
1986, 25, 3917Ϫ3925.
L. Reven, J. Mol. Catal. 1994, 86, 447Ϫ477.
J. Blumel, Inorg. Chem. 1994, 33, 5050Ϫ5056.
K. D. Behringer, J. Blumel, Chem. Commun. 1996, 653Ϫ654.
[16]
[17]
[18]
[45]
[46]
[19]
[20]
[47]
[48]
S. Hayashi, K. Hayamizu, Bull. Chem. Soc., Jpn. 1989, 62,
2429Ϫ2430.
K. Okuro, M. Furuune, M. Enna, M. Miura, M. Nomura, J.
Org. Chem. 1993, 58, 4716Ϫ4721.
[21]
[22]
[23]
[24]
Received July 6, 2004
Early View Article
Published Online December 2, 2004
Eur. J. Inorg. Chem. 2005, 496Ϫ503
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
503