Bipolar electrochemistry is a concept with a quite long history, but has only very recently revealed its virtues in the field nanoscience. It allows controlled surface modification at the micro- and nanoscale, with very original applications ranging from analytical chemistry to material science. Here we review some of the latest achievements in the field and explain why this straight forward concept leads to a very attractive tool for shaping the micro- and nanoworld.

The Past
The concept of bipolar electrochemistry is known for quite a long time, but it has never crossed the frontier of a few industrial applications, that have been developed in the 1960ies. Traditional examples of bipolar electrochemistry can be found under the name of fluidized bed electrodes for applications such as electrosynthesis, water splitting and also for increasing fuel cell performances [1,2]. The very simple concept behind these processes is based on the fact that a bipolar redox behavior is induced on a substrate under the influence of an external electric field [3].

Let's consider a conducting object immersed in a solution that is placed in a strong electric field between two electrodes (Figure 1). Between the two sides of the object a polarization ΔV occurs, that is proportional to the electric field E/L and the characteristic dimensions of the object r.

ΔV= 2 E r/L (1)

When this polarization is strong enough, redox reactions can be carried out at the opposite ends of the object. The negatively polarized side can be for example the site of metal ion reduction, leading to the formation of a metal cluster, whereas at the positively polarized side an electropolymerization can lead to the formation of a layer of conducting polymer.

Different advantages become immediately clear when looking at this concept: first of all it allows the controlled modification of an object using an electrochemical process without any physical connection of the object to an electrode. This means that in principle the same processes can be triggered on thousands or millions of objects simultaneously, which makes the approach very attractive in highly parallelized devices like integrated microelectrode arrays [4].

A second obvious argument in favor of this technique is the fact that it is one of the rare approaches that allows breaking the symmetry in a modification process without using a surface or an interface. It is therefore a straightforward synthetic way to dissymmetric particles obtained via a true bulk process. These dissymmetric objects, also named Janus particles, after the roman God who had two different faces, are of great importance for studying fundamental and applied issues in micro- and nanotechnology, like directional self-assembly, nanoelectronics, photosplitting of water and electronic paper.

In principle the initial objects can be made out of any kind of conductive material with any characteristic dimensions and geometry. However, as shown by equation (1), a smaller object will require a higher external electric field in order to be polarized enough to induce redox reactions, and this is an intrinsic problem when modifications of nanometer sized objects should be carried out. One can easily calculate that in this case typical external electric fields of the order of MV/m are needed in order to achieve a dissymmetric modification, values that seem barely compatible with normal laboratory conditions. Therefore most of the work in the past using bipolar electrochemistry has been carried out with rather big objects, in the best case with dimensions in the submicrometer range [5].

The Present
In order to circumvent the problems associated with the above mentioned high voltages, like bubble formation due to solvent decomposition and rotation of the objects during the modification, it is possible to use a set-up that separates the anodic and cathodic compartment via a capillary (Figure 2). While the objects are driven through the capillary due to the intrinsic electroosmotic flow, they get at the same time modified if the necessary ingredients like metal salts or monomers are present in the solution. At the outlet of the capillary the Janus particles can be easily collected and then characterized. This Capillary Assisted Bipolar ElectroDeposition (CABED process) allows indeed the selective modification of nanometer sized objects, because overall voltages of up to 50 kV, corresponding to electric fields almost in the MV/m range, can be applied, analogue to what is routinely achieved with a classic capillary electrophoresis equipment [6].

This set-up led for the first time to the site selective generation of metal clusters on carbon nanotubes (Figure 3).
The CABED process can also be used for the localized modification of other objects such as micrometer sized carbon tubes, as illustrated on the right side of Figure 3 for the toposelective deposition of platinum [7]. Other metals that have been successfully tested include copper and nickel [8]. Because nickel has a quite negative deposition potential, one can conclude that a large variety of metals is accessible for modifying this kind of objects using the same approach. As the condition for successful modification is that the object needs to be a good conductor in order to allow electron transfer from one side to the other, one can for example imagine to use nickel modified carbon nanotubes to collect only the tubes which are good conductors with the help of a magnet, as the semiconducting or insulating tubes will not be modified.