Technologies for producing electric contacts on the illuminated side of solar cells are based on chemical processes. Silver technologies are widely used for manufacturing crystalline silicon solar cells. The role of small particles in solar cells was described previously (Hitz, 2007; Pillai, 2007; Han, 2007; Johnson, 2007). The introduction of nanoparticles into pores of photon absorbers increases their efficiency. In our experiments copper microclusters were chemically introduced into pores of a silver contact. They changed the electrical properties of the contact: dark current, which is unknown for metals, was detected.
In the experiments, we used 125 x*125-mm commercial crystalline silicon wafers Si<P>/SiNx (70 nm)/Si<B> with a silver contact on the illuminated side. The silver contact was porous silver strips 10-20 pm thick and 120-130 pm wide on the silicon surface. The diameter of pores in a contact strip reached 1pm. The initial material of the contact was a silver paste (Dupont), which was applied to the silicon surface through a tungsten screen mask. After drying, organic components of the paste were burned out in an inert atmosphere at 820-960° C. Simultaneously, silver was burned in into silicon through a 70- nm-thick silicon nitride layer. After cooling in air, the wafer was immersed in a copper salt solution under the action of an external potential difference; then, the wafer was washed with distilled water and dried with compressed nitrogen until visible removal of water from the surface of the solar cell (Laptev & Khlyap, 2008).
The crystal structure of the metal phases was studied by grazing incidence X-ray diffraction. A 1-pm-thick copper layer on the silver surface has a face-centered cubic lattice with space group Fm3m. The morphology of the surface of the solar cell and the contact strips before and after copper deposition was investigated with a KEYENCE-5000 3D optical microscope. Fig. 15 presents the result of computer processing of images of layer-by-layer optical scanning of the surface after copper deposition.
Fig. 15. Contact strip morphology. Scanning area 430 x 580 pm2; magnification 5000x.
The copper deposition onto the silver strips did not change the shape and profile of the contact, which was a regular sequence of bulges and compressions of the contact strip. The differences in height and width reached 5 pm. In some cases, thin copper layers caused slight compression of the contact in height. The profiles of the contacts were studied using computer programs of the optical microscope. It was found that copper layers to 1 pm in thickness on the silver contact could cause a decrease in the strip height by up to 10%.
The chemical composition of the contact and the depth distribution of copper were investigated by energy dispersive X-ray analysis, secondary ion mass spectrometry, and X – ray photoelectron spectroscopy. The amount of copper in silver pores was found to decrease with depth in the contact. Copper was found at the silicon-silver interface. No copper diffusion into silicon was detected.
The resistivity of the contacts was measured at room temperature with a Keithley 236 source-measure unit. Two measuring probes were placed on the contact strips at a distance of 8 mm from each other. A probe was a tungsten needle with a tip diameter of 120 pm. The measurements were made on two samples in a box with black walls and a sunlight simulator. Fig. 16 presents the results of the experiments.
Line 1 is the current-voltage diagram for the initial silver contact strip on the silicon wafer surface. The other lines are the current-voltage diagrams for the contacts after copper deposition. All the lines confirm the metallic conductance of the contact strips. The current – voltage diagrams for the contacts with copper clusters differ by the fact that they do not pass through the origin of coordinates for both forward and reverse currents. A current through a metal in the absence of an external electric field is has not been observed. In our experiment, the light currents were 450 pA in the contact where copper clusters were only in silver pores and 900 pA in the contact where copper clusters were both in silver pores and on the silver surface.
Fig. 16. Electrical properties of (1) a silver contact strip, (2) a contact strip with copper clusters in silver pores, and (3) a strip with a copper layer on the surface and copper clusters in silver pores.
It is worth noting that the electric current in the absence of an external electric field continued to flow through these samples after the sunlight simulator was switched off. The light and dark currents in the contact strips are presented in Fig. 17. It is seen that the generation of charge carriers in the dark at zero applied bias is constant throughout the experiment time. The dark current in the silver contact is caused by the charge carrier generation in the contact itself. The source of dark-current charge carriers are copper clusters in silver pores and on the silver surface.
Fig. 17. Time dependence of the (1) dark and (2) light currents at zero applied bias in contact strips with copper clusters in silver pores.
The current in the silver contact with copper clusters while illuminating the solar cell is caused by the generation of charge carriers in the semiconductor part of the silicon wafer. The number of charge carriers generated in the p-n junction is two orders of magnitude larger than the number of charge carriers in copper clusters since the light current is so larger than the dark current (Fig. 17).
The copper deposition onto silver does not lead to the formation of a silver-copper solid solution. The contact of the crystal structures gives rise to an electric potential difference. This is insufficient for generation of current carriers.
However, the contact of the copper and silver crystal structures causes compression of the metal strip and can decrease the metal work function of copper clusters.
We consider that the charge carrier generation in the dark by copper clusters in the contact strip as a component of the solar cell is caused by the deformation of the strip. It is known (Albert & Chudnovsky, 2008), that deformation of metal cluster structures can induce high – temperature superconductivity. Therefore, it is necessary to investigate the behavior of the studied samples in a magnetic field.
Solar energy conversion is widely used in electric power generation. Its efficiency in domestic and industrial plants depends on the quality of components (Slaoui A & Collins, 2007). Discovered in this work, the dark current in the silver contact on the illuminated side of a silicon solar cell generates electricity in amount of up to 5% of the rated value in the absence of sunlight. Therefore, the efficiency of solar energy conversion plants with copper – silver contacts is higher even at the same efficiency of the semiconductor part of the solar cell.