Indeed, in Figure 3b, on the right axis, the variation of O S wit

Indeed, in Figure 3b, on the right axis, the variation of O S with thickness in the c-Ge QW is reported, as TH-302 cost calculated in the 5- to 35-nm thickness range by Kuo and Li, using a 2D exciton

model and infinite barrier [6]. The good agreement between measured B and calculated O S is the experimental confirmation that the enhanced absorption efficiency observed at room temperature in a-Ge QWs is actually due to the excitonic effect. The inset of Figure 3b evidences the linear correlation between B (measured at 5, 12, and 30 nm) and the expected O S (for those thicknesses), allowing for the estimation of the factor of proportionality (γ = B/O S , which accounts for the absorption efficiency normalized to the oscillator strength). Thus, a proper modeling applied to light absorption measurements at room temperature allowed to quantify the extent of size effect in a-Ge QWs and to disentangle the oscillator strength increase and the bandgap widening in these structures. In order to test if photogenerated carriers in a-Ge QWs can be separated and collected through the action of an external electric field, we realized Temsirolimus clinical trial a photodetector device, as illustrated in the drawing of Figure 4, and performed transversal current density versus voltage (J-V) measurements in dark and under white

light illumination conditions. Figure 4 reports the J-V curves for samples with 12-nm (Figure 4a) or 2-nm (Figure 4b) a-Ge QW. In dark conditions, both the MIS devices (biased as shown in the drawing) have similar behavior in forward and reverse biases. Most of the applied voltage is dropped across the dielectric (SiO2) stacks, while the QW thickness slightly lowers the dark current density (J dark) in the thicker sample (offering a more resistive path). Upon white light illumination, J-V values remain largely unaffected in the forward bias, while an increase of the current density (J light) occurs for the thicker samples in the reverse bias

regime. In particular, for a negative bias of −3 V, the net photocurrent (J light − J dark) increases from 1 to 12 μA/cm2 going from 2 to 12 nm of QW thickness. The net photocurrent is due to the electron-hole pairs photogenerated in the QW and in the substrate (n-Si). As the device is reverse biased, electrons are pushed to the substrate and holes to PAK6 the transparent electrode. It should be noted that by increasing the Ge QW thickness, the contribution of the substrate to the net photocurrent shrinks. In fact, the photogeneration of electron-hole pairs in the substrate decreases because of the light absorbed in the QW, and the carrier collection lowers because of the higher resistance. By comparing the images in Figure 4a,b, we can appreciate the role of the a-Ge film, as the MIS devices differ only for the QW thickness. The higher net photocurrent measured in the thicker QW gives a clear evidence of a positive photoconductivity effect within a-Ge QWs.

Comments are closed.