Controlled deposition of nanoparticles at the liquid-liquid interface

Article abstract of DOI:10.1039/b208275f

We report a novel method for the preparation of well-defined metallic nanoparticles, which is illustrated with the technologically important example of palladium nanoparticles dispersed in γ-alumina.

Full text of DOI:10.1039/b208275f

Controlled deposition of nanoparticles at the liquid–liquid interface
Mark Platt,a Robert A. W. Dryfe*a and Edward P. L. Robertsb
a Department of Chemistry, University of Manchester Institute of Science & Technology, PO Box 88,
Manchester, UK M60 1QD. E-mail: robert.dryfe@umist.ac.uk; Fax: +44 161 200 4559; Tel: +44 161 200
4522
b Department of Chemical Engineering, University of Manchester Institute of Science & Technology, PO
Box 88, Manchester, UK M60 1QD
Received (in Cambridge, UK) 27th August 2002, Accepted 10th September 2002
First published as an Advance Article on the web 25th September 2002
We report a novel method for the preparation of well-
defined metallic nanoparticles, which is illustrated with the
technologically important example of palladium nano-
particles dispersed in g-alumina.
of Pd was confirmed by energy dispersive X-ray analysis (EDX,
see Table 1). The structure of the alumina template is generally
apparent, indicating that Pd deposits are growing as discrete
nanoparticles.
The dark patches observed in Fig. 2(a)–(c), represent the
increasingly large growths of Pd nanoparticles. The particles do
not coalesce, even at high Pd coverage, indicating that the
alumina membrane serves as an effective template to control
deposition at the organic/water interface. The white patches
visible in Fig. 2(a)–(c) are due to localised sample charging
caused by the presence of precipitates of aluminium oxide.9 The
Pd nanoparticles thus formed are flush with the alumina
surround and have a high surface area with a maximum
dimension determined by the pore diameter employed (in this
case, 100 nm). To confirm the templated-growth mechanism of
the membrane-supported Pd, the alumina surround was re-
moved by dissolution over 72 h in 1 M HCl. The liberated Pd
particles were washed and filtered, with SEM analysis per-
formed subsequently (Fig. 2(d)). The images show that some
particle aggregation has occurred following template dissolu-
tion,10 however the particle morphology retains some spherical
structure, with an approximate diameter of 100 nm, indicative
We present a novel method for the stable and controlled
electrochemical deposition of metallic palladium within a
porous g-alumina substrate, a system of relevance in catalysis
and gas separation.1,2 The method exploits electrodeposition at
the template-modified interface between immiscible liquids,
where the g-alumina acts as the template, to control palladium
formation.
Electrochemical deposition at organic/water interfaces in the
absence of templates has been reported previously.3–5 This
process is of intrinsic interest due to the lack of solid substrate,
and associated defect sites, where the energetic barrier to
nucleation is assumed to be lowest.6 However, understanding of
the mechanism of electrochemical deposition at the organic/
water interface has been hampered by the tendency of the
deposited nanoparticles to aggregate.5 The method described
here replaces the metallic substrate previously employed for
template-controlled electrochemical deposition7,8 with an im-
miscible organic solvent. The g-alumina membranes were used
to separate the aqueous and organic phases of the electro-
chemical cell, with the aqueous phase forming the lower half of
the cell. The organic/water interface is localized at the mouth of
the template because of the membrane’s hydrophilic proper-
ties.† Application of a potential difference across the resultant
array of pore-sized organic/water interfaces caused controlled
deposition of Pd at the mouth of the alumina pore. An interfacial
reduction of aqueous phase tetrachloropalladinate to metallic
palladium is induced by the potential difference,5 with organic
phase butylferrocene employed as the electron donor.‡ Pd is
thus controllably dispersed within the pores of the alumina
template. By contrast with previous template approaches,7,8 the
dispersed nanoparticles are not formed on an underlying
metallic substrate, thereby avoiding the requirement for sub-
strate removal.
Both voltammetry and chronoamperometry of the electrified
alumina-templated liquid–liquid interface displayed responses
characteristic of nucleation (Fig. 1),5 with an irreversible
process at +0.7 V, which corresponds to the deposition of Pd.
The nucleation loops are no longer observed after approx-
imately six successive cyclic voltammetric scans, indicating
that the current on consecutive scans contains a progressively
decreasing contribution from nucleation. Fig. 2(a) shows a high
resolution scanning electron microscope (SEM) image of an
alumina membrane, following an initial (5 min) electrochemical
deposition of Pd particles at the templated 1,2-dichloroethane
(DCE)/water interface. The nominal diameter of the alumina
pores was 100 nm, the cell was configured with the organic
phase above the alumina membrane, with the aqueous phase on
the underside. Following this initial deposition, the SEM
indicates that around 5% of the pores contain detectable
amounts of Pd. Prolonged electrolysis of the DCE/water
interface leads to deposition within an increasingly large
fraction of the available pores (Fig. 2(b) and (c)). The presence
Fig. 1 (a) Cyclic voltammetry of Pd deposition cell at a scan rate of 0.1 V
s21, aqueous phase [PdCl422] of 1 3 1023 M, (A) shows the response in the
absence of butylferrocene, (B) is the response with organic phase
[butylferrocene] of 5 3 1024 M. The arrows indicate the nucleation loop
observed. (b) Successive potential steps from 20.45 V to (A) 20.9 V, (B)
22
20.8 V and (C) 20.75 V, where the cell contains aqueous phase [PdCl4
]
(1 3 1023 M) and organic phase butylferrocene (5 3 1024 M). The organic
phase was placed above the aqueous phase in these cases.
2324
CHEM. COMMUN., 2002, 2324–2325
This journal is © The Royal Society of Chemistry 2002

View Article Online
of the constraint over growth imparted by the template.
developed offers simpler control over particle morphology than
Electrochemical deposition at the organic/water interface
offers a unique way to observe the relationship between the
driving force for particle growth (here, the applied potential)
and the form of the resultant particles. By contrast, the initial
stages of electrodeposition on solid substrates are governed by
the structure of the substrate, leading to phenomena such as
under-potential deposition and epitaxial growth.11–13 The
method pioneered here offers a route to the fabrication of
metallic nanostructures that does not require an electrical
contact to be made to the nanostructure, since particle growth is
at the interface between two electrolyte solutions. The method
colloidal growth methods, hence it represents an interesting
intermediate case between classical electrochemical and chem-
ical routes to the growth of metallic particles.6,14,15
We are currently investigating the generalisation of our
method to the production of other nanostructured materials, as
well as exploring the detailed influence of applied potential on
Pd growth. In due course, we envisage important applications,
for example in hydrogenation catalysis and separations,1,2 for
the nanoparticle ensembles generated via this technique.
The authors thank the EPSRC for financial support and Mr I.
Brough (Manchester Materials Science Centre) for assistance
with SEM and EDX.
Notes and references
† Membranes of g-alumina, with nominal pore diameters of 100 nm and
thicknesses of 60 mm (Whatman International Ltd, Maidstone, UK) were
sealed to a 5 mm i.d. glass tube with silicone sealant.
‡ Electrochemical experiments were performed using an EG&G Model 273
potentiostat (Princeton Applied Research Corp., Princeton NJ, USA)
operating in four-electrode mode. The aqueous phase comprised of 1 3
1023 M ammonium tetrachloropalladate (Fluka, Gillingham, UK), 5 3
1022 M lithium chloride and 5 3 1022 M lithium sulfate dissolved in water
obtained from a Milli-Q purification system. The organic solvent,
1,2-dichloroethane (HPLC grade, Lancaster Synthesis, Lancaster, UK) was
used with an organic electrolyte, bis(triphenylphosphoranylidene)ammon-
ium tetrakis(pentafluoro)phenylborate, prepared in house, at a concentra-
tion of 5 3 1023 M. During deposition experiments butylferrocene
(Lancaster, 98%) was added to the DCE solution in concentrations of either
5 3 1024 or 1 3 1023 M. Ag/AgCl reference electrodes were employed:
potentials are quoted with respect to these electrodes.
Fig. 2 (a) SEM image of Anopore g-alumina membrane, after deposition of
Pd, at +0.7 V using the cell described‡ for 5 min. (b) SEM image of g-
alumina membrane (nominal pore size 100 nm), after deposition of Pd for
1 h at a potential of 0.95 V. (c) SEM image of alumina membrane following
deposition for 1 h at a potential of 0.95 V, as in (a), although the
concentration of organic phase butylferrocene was increased (to 1 3 1023
M). (d) SEM image of Pd particles electro-deposited, following dissolution
of the alumina membrane in 1 M HCl. The liberated Pd particles were
retained by filtration with a polyester filter (Poretics, Livermore, CA, USA),
with a mean pore diameter of 100 nm. The scale bars in the SEM images
correspond to 0.5 mm for (a), (b) and (d), and 1 mm in the case of (c).
1 S. I. Niwa, M. Eswaramoorthy, J. Nair, A. Raj, N. Itoh, H. Hoji, T.
Namba and F. Mizukami, Science, 2002, 295, 105.
2 X. L. Pan, G. X. Xiong, S. S. Sheng, N. Stroh and H. Brunner, Chem.
Commun., 2001, 2536.
3 M. Guainazzi, G. Silvestri and G. Serravalle, J. Chem. Soc., Chem.
Commun., 1975, 200.
4 Y. Cheng and D. J. Schiffrin, J. Chem. Soc., Faraday Trans., 1996, 92,
3865.
5 C. Johans, P. Liljeroth and K. Kontturi, Phys. Chem. Chem. Phys., 2002,
4, 1067.
6 R. M. Penner, J. Phys. Chem. B, 2002, 106, 3339.
7 C. R. Martin, Science, 1994, 266, 1961.
8 G. L. Hornyak, C. J. Patrissi and C. R Martin, J. Phys. Chem. B, 1997,
101, 1548.
Table 1 Composition, by wt% and atom% of the membrane shown in Fig.
2(a), as determined by EDX
9 C. J. Miller and M. Majda, J. Electroanal. Chem., 1986, 207, 49.
10 C. Johans, K. Kontturi and D. J. Schiffrin, J. Electroanal. Chem., 2002,
526, 29.
11 M. Paunovic and M. Schlesinger, Fundamentals of Electrochemical
Deposition, Wiley, New York, 1998.
12 G. A. Somorjai, Surf. Sci., 1995, 335, 10.
13 M. Gierer, H. Over, H. Bludau and G. Ertl, Phys. Rev. B, 1995, 52,
2927.
14 C. C. Wang, D. H. Chen and T. C. Huang, Colloid Surf. A: Physicochem.
Eng. Asp., 2001, 189, 145.
Sample
Element
Wt%
Atom%
Blank membrane
O
Al
P
29.86
63.07
3.04
42.15
52.79
2.22
2.84
31.67
57.79
2.31
2.94
5.3
S
4.04
Palladium membrane
O
Al
P
S
Pd
18.12
55.78
2.56
3.37
20.17
15 B. Ludolph, M. A. Malim, P. O’Brien and N. Revaprasadu, Chem.
Commun., 1998, 1849.
CHEM. COMMUN., 2002, 2324–2325
2325