Search

Some crucial limiting process steps of ITO/(n)a-Si:H/(p)c-Si solar cell manufacturing were investigated. The influence of amorphous silicon deposition temperature on the efficiency of heterojunction (HJ) solar cells as well as ITO deposition on the properties of ITO/a-Si:H/c-Si interfaces were studied. Raman spectroscopy, X-ray photoelectron spectroscopy (XPS) and SIMS measurements were employed for the investigations. It was observed that the decrease in a-Si:H deposition temperature from 230 down to 100°C leads to a reduction in solar cell efficiencies to only 1%. It was found that the ITO deposition process itself can significantly modify the ITO/a-Si:H/c-Si interfaces due to the interaction of oxygen, and In and Sn atoms with silicon at the initial stage. It was concluded that this modification has a more significant impact on the quality of the HJ solar cells than on that of the a-Si:H layer.

The properties of the thermosetting polymer mr-I-9000, developed and used for nanoimprint lithography (NIL), were investigated by means of X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR). The spectra of mr-I-9000 were compared with those of PMMA and show the presence of aromatic rings and a high amount of carbon–carbon and carbon–hydrogen bonds. As a consequence, the mask selectivity of mr-I-9000 in an oxygen and tetrafluoromethane reactive ion etching processes is higher than that for PMMA and slightly different for the cross-linked and non-cross-linked state.

We summarize flow and filling effects detected during hot embossing lithography (HEL), which reflect the visco-elastic properties of the polymers as well as the specific processing conditions and stamp pattern configurations. In particular a delayed elastic recovery is reported, which is found in larger elevated stamp regions. The consequences of these effects for mask definition by HEL are discussed, and strategies for wafer-scale imprint are de-duced. Successful imprint of 4' wafer is demonstrated under adequate processing conditions.

Mask definition was performed by use of nanoimprint lithography and subsequent reactive ion etching in an oxygen plasma. Polystyrene was chosen as a polymer mask material. Different features ranging from 400 nm up to 4 micrometers were produced in the polymer layer by nanoimprint. Optimization of the residual layer removal process in oxygen RIE was performed at different pressures and self-bias voltages. Low pressure and high bias voltage are required for high quality mask definition.

Replication of the e-beam master into thermosetting polymers was investigated for low cost working stamp fabrication for nanoimprint lithography (NIL). Negative as well as positive replicas from a master were fabricated in a thermal replication step from duroplastic material. Without any further treatment these working stamps were able to imprint thermoplastic polymers. To demonstrate the procedure we used the custom polymers mr-I-8130 and mr-I-9030 as well as PMMA for comparison. All replications were done in a NIL process. The replication quality is excellent when a suitable imprint procedure is used.