|
Room Temperature Coulomb Blockade in Arrays of 28 kDa Au Nanocrystals
We have performed local complementary mapping of the topography and charge transport characteristics of arrays of ligand-protected 28 kDa gold nanocrystals via combined conducting-probe atomic force microscopy
(CP-AFM) and displacement-voltage (z-V) spectroscopy. Figure 1
shows a schematic of the z-V measurement process.
|
|
|
Figure 1. Schematic representation of a local
z-V measurement on a small array of ligand-protected 28 kDa Au nanocrystals located between a
metallised AFM probe and a planar gold substrate.
(a) Left: 1-D representation of a single conducting path; a simplified energy diagram for this path at a substrate bias voltage
(V) below the array Coulomb blockade voltage threshold (V <
VT). Coulomb blockade has been lifted across all but one of the nanocrystals; the equivalent circuit for this path.
Right: Cross-sectional schematic of the probe-array-substrate electrical nanocontact.
(b) Situation for V ³ VT, where the lifting of the local Coulomb blockade across each of the nanocrystals in the path results in a sharp increase in conductance, causing the probe to retract in order to maintain the preset current.
|
Figure 2. (a) Tapping mode AFM topography of a layer of 28 kDa Au nanocrystals
(diameter d = 1.65 nm) drop-cast on a Au substrate measured before acquisition of probe displacement versus bias voltage
(z-V) data.
(b) Same region imaged following acquisition of z-V data at a regular grid of locations (with pitch ~ 750 nm).
Acquisition of z-V data does not appear to affect the nanocrystal layer topography.
(c) Negative and positive bias z-V curves (blue curve) acquired at the same location on the 28 kDa nanocrystal film. Clear Coulomb blockade voltage thresholds are observed at both polarities. No thresholds are observed in
z-V data acquired on bare Au substrates (red curve) or in data measured for drop cast layers of larger
d = 6 nm Au nanocrystals (green curve).
|
The sharp thresholds observed at both bias polarities for the 28 kDa nanocrystal array
z-V measurement data shown in Figure 2c are well described by the model presented in Figure 1. If we consider the positive sweep, at bias voltages below the measured voltage threshold
(V < VT+), each conducting path through the nanocrystal array contains at least one junction under Coulomb blockade so the probe displacement is minimal and similar to the observed behavior for measurements on bare gold substrates; see Figure 2c. When the substrate bias is sufficiently large
(V = VT+), the Coulomb blockade is lifted, the conductance increases sharply and the probe retracts to maintain the constant preset current. Above the threshold
(V ³ VT+), many additional conduction paths can open and the probe retracts further in order to maintain the setpoint current. A similar situation holds for the negative sweep, where the probe displacement is minimal for
|V| < |VT-| and the Coulomb blockade is lifted at V =
VT-.
For each pair of z-V measurements at a given location, we define the difference in energy between the positive and negative Coulomb blockade threshold voltages to be
Ediff = e(VT+- VT-) = 1.4 eV for the data shown in Figure 2c. For sequential Coulomb blockade,
Ediff = 2 N Ec, where N is the integer number of nanocrystals in the shortest conducting path(s) through the portion of the array between the probe apex and the substrate and
Ec is the single electron charging energy for a 28 kDa nanocrystal in an array
(Ec ~ 270 meV as measured by differential pulse voltammetry). The measured value of
Ediff for the data shown in Figure 2c therefore suggests N ≈ 3 nanocrystals in the shortest path(s) through the portion of the nanocrystal array between the apex of the conducting probe and the
substrate. The probe apex area ~ 25 x 25 nm2, corresponding to an array with lateral dimensions of ~ 9
x 9 nanocrystals. The conducting AFM probe thus acts as an electrical nanocontact.
The similarity in the topography data measured before (Figure
2a) and after (Figure 2b) acquisition of z-V data at a grid of
measurement locations within this region indicates that the nanocrystals reflow in a fluid-like manner as the
probe retracts during z-V measurements at each location, enabling reproducible formation of probe-array-substrate nanocontacts.
Publications
O’Brien, G.A.; Quinn, A.J.; Biancardo, M.; Preece, J.A.; Bignozzi, C. A. & Redmond, G.
Making Electrical Nanocontacts to Nanocrystal Assemblies: Mapping of Room Temperature
Coulomb Blockade Thresholds in Arrays of 28 kDa Nanocrystal Gold Molecules.
Small 2,
261 (2006).
|
