One of many factors which can influence the electrochemical behaviour of polycrystalline boron doped diamond (pBDD) electrodes is the boron doping density. At boron dopant levels of ca. 2 x 1020 atoms cm-3 the material undergoes a transition from p-type semiconducting to semimetallic. However, the surface of pBDD is heterogeneous due to the presence of different crystallographic orientations. Boron is more readily incorporated (by a factor of approximately 10) into (111) growth sectors than (100) sectors. Hence, even if high boron doping levels are employed in the chemical vapour deposition (CVD) process, different grains may show different electrical and electrochemical activities due to this differential boron uptake (spanning insulating to semiconducting to semimetallic). We have examined the electrochemical and electrical characteristics of thick films (ca. 500 µm) of oxygen terminated pBDD. polished to nm level roughness. The pBDD was grown and polished by Element Six with an average boron dopant density of ca. 5 x 1020 atoms cm-3.
Cyclic Voltammetric Studies
Cyclic voltammetric studies of the reduction of 0.1 mM and 10 mM Ru(NH3)63+ (in 0.1 M KCl) at various scan rates were recorded (Figure 1). These studies were performed at potentials substantially negative of the flat-band potential for pBDD. This means that for any regions of the pBDD surface that are nondegenerately boron-doped (non metal like), electron transfer occurs with the pBDD in the depletion region. Increasing peak-peak separation was observed as the concentration of redox mediator was increased (as well as with scan rate). Although the rate of electron transfer should be independent of redox species concentration, the effect of a limited number of charge carriers in the pBDD depletion layer may lead to the observed kinetic effects. Alternatively, the CV response shown in (b) may also result from resistance effects in the diamond film.
Figure 1. Cyclic voltammograms for the reduction of (a) 0.1 mM Ru(NH3)63+ and (b) 10 mM Ru(NH3)63+ in 0.1 M KCl at pBDD at scan rates of 2, 5, 10, 20, 50, 100, 200 and 500 mV s-1.
Conducting Atomic Force Microscopy
Conducting atomic force microscopy (C-AFM) images of pBDD were recorded in air. The height and conductivity of the pBDD were simultaneously measured (Figures 2 and 3). The topography images show a root-mean-square roughness of ~ 1 nm. This low surface roughness is vital for effective C-AFM.
Conductivity readings were initally taken with a tip potential of -5 V with a 10 MΩ current-limiting resistor in series (Figure 2b). Two types of regions are clearly present on the pBDD surface: the first being ‘high’ conductivity regions (white) where the current is ~ 500 nA. The second are ‘lower’ conductivity regions (black) where the current is ~ 250 nA. The lower conductivity regions have obvious variations in them, unlike the higher conductivity regions which show little structure. This is due to the fact that at all points the resistance is significantly less than the 10 M current-limited resistor used.
Figure 2. Simultaneously recorded conducting-AFM images (125 x 125 µm) of (a) height and (b) conductivity of the pBDD sample, recorded in air with resistance in series.
C-AFM images were also taken with a tip potential of +50 mV, with no resistor in series (Figure 3b). Here, the lower conductivity regions are white and correspond to currents of less than 50 nA. Obvious variations within the higher conductivity regions can be seen, with resistances ranging from 10 to 100 kΩ. There is a very sharp transition from the low to high conductivity regions, indicating that lateral conductivity on the surface in this case is insignificant. Also, no evidence of enhanced conductivity at grain boundaries was found.
Figure 3. Simultaneously recorded conducting-AFM images (50 x 50 µm) of (a) height and (b) conductivity of the pBDD sample, recorded in air with no resistor in series.
Current-voltage curves of the two regions were also taken and plotted on log-log plots (Figure 4). A difference in conductivity of over 2 orders of magnitude can be seen.
Figure 4. Current-voltage curves taken in (a) a high conductivity region and (b) a low conductivity region, recorded inair with no resistor in series. (c) Comparison of current-voltage curves from the two regions on a log-log plot.
Scanning Electrochemical Microscopy
Scanning electrochemical microscopy (SECM) was used in a substrate generation-tip collection mode, with a 25 um diameter tip tip scanned at a constant height of 3 µm above the sample. The tip was held at 0.0 V to detect Ru(NH3)62+ at a diffusion-controlled rate. 500 x 500 µm scans were run for the tip collection of electrogenerated Ru(NH3)62+ from the pBDD surface. The pBDD was held at (a) -0.23 and (b) -0.3 V versus AgQRE to drive the reduction of Ru(NH3)63+ at different rates on the surface (Figure 5).
At all driving potentials the entire surface is active, however, electrogeneration of Ru(NH3)62+ clearly occurs at different rates across the surface. When the driving potential is increased, tip currents are larger, indicating a faster rate at the substrate. A similar pattern of spatially dependent activity is still observed. These patterns of activity correspond well with the C-AFM data, but at lower resolution. On a metal, a positive normalised tip current of ca. 4 would be expected at diffusion-controlled rates. Thus, some of the areas on the pBDD surface demonstrate ‘metal-like’ behaviour. These are the areas corresponding to dopant levels in excess of 1020 atoms cm-3. The lower currents in the SECM images most likely correspond to domains where the charge carrier concentrations or mobility is lower. As electron transfer still occurs, it suggests a thin depletion region in the lower conductivity domains.
Figure 5. 500 x 500 µm SECM tip collection scans for the collection of electrogenerated Ru(NH3)62+. The substrate was held at (a) -0.23 and (b) -0.3 V to drive the reduction of Ru(NH3)62+ at differnt rates.
- N. R. Wilson, S. L. Clewes, M. E. Newton, P. R. Unwin, J. V. Macpherson, J. Phys. Chem. B 2006, 110(11), 5639-5646J. Phys. Chem. B 2006, 110(11), 5639-5646.