Radio frequency (RF) biasing induced by a second plasma source at the substrate is applied to low-temperature sputtering processes for indium tin oxide (ITO) and titanium nitride (TiN) thin films. Investigations on crystal structure and surface morphology show that RF-biased substrate plasma processes result in a changed growth regime with different grain sizes and orientations than those produced by processes without a substrate bias. The influence of the RF bias is shown comparatively for reactive RF-sputtered ITO and reactive direct-current-sputtered TiN. The ITO layers exhibit an improved electrical resistivity of 0.5 mΩ cm and an optical absorption coefficient of 0.5 × 104 cm−1 without substrate heating. Room-temperature sputtered TiN layers are deposited that possess a resistivity (0.1 mΩ cm) of 3 orders of magnitude lower than, and a density (5.4 g/cm3) up to 45% greater than, those obtained from layers grown using the standard process without a substrate plasma.

Dielectric nanomaterials are emerging as key components in today's highly efficient silicon solar cells. The most successful materials are SiO2, SiNx: H and Al2O3 due to their excellent material properties for surface passivation and light management. Dielectric passivation layers ensure high level of chemical passivation by effectively reducing the density of silicon surface states, which could act as recombination centers for photo-generated charge carriers. Additionally, these materials provide strong field-effect passivation by a high density of intrinsic fixed charges. However, novel solar cell concepts demand for dielectric nanomaterials with additional functionalities, such as electrical conductivity, simultaneous applicability on n-and p-type substrates or compatibility with low-thermal budget processing. Recent developments show that these functionalities can be realized in multi-oxide nanolaminates including HfO2 or TiO2 sublayers. Additionally, alternative deposition techniques and innovative processes, such as flash light annealing, become available to reduce process complexity and thermal budget. This chapter reviews the theory and application of dielectric nanomaterials in today's solar cells and gives an outlook on promising solutions in future solar cells.

In Al2O3-based passivation layers, the formation of fixed charges and trap sites can be strongly influenced by small modifications in the stack layout. Fixed and trapped charge densities are characterized with capacitance voltage profiling and trap spectroscopy by charge injection and sensing, respectively. Al2O3 layers are grown by atomic layer deposition with very thin (∼1 nm) SiO2 or HfO2 interlayers or interface layers. In SiO2/Al2O3 and HfO2/Al2O3 stacks, both fixed charges and trap sites are reduced by at least a factor of 5 compared with the value measured in pure Al2O3. In Al2O3/SiO2/Al2O3 or Al2O3/HfO2/Al2O3 stacks, very high total charge densities of up to 9 × 1012 cm−2 are achieved. These charge densities are described as functions of electrical stress voltage, time, and the Al2O3 layer thickness between silicon and the HfO2 or the SiO2 interlayer. Despite the strong variation of trap sites, all stacks reach very good effective carrier lifetimes of up to 8 and 20 ms on p- and n-type silicon substrates, respectively. Controlling the trap sites in Al2O3 layers opens the possibility to engineer the field-effect passivation in the solar cells.

A silicon heterojunction solar cell based on amorphous and crystalline silicon is combined with the metal wrap through technology. In this novel solar cell concept one critical process is the via hole conditioning. Raman measurements reveal that the amorphous silicon emitter layer hardly penetrates the via holes and that thereby the via surface is not fully covered. In the conventional process sequence with via hole formation prior to wet chemical cleaning, the effective carrier lifetime is reduced by about 50 % in the vicinity of the via hole. An improved process sequence is presented, which bases on via hole formation after the thin film depositions. In this sequence, the via hole formation process is crucial for the via surface passivation. The passivation remains poor when applying a 1064 nm laser process. However, very good surface passivation is achieved with a 532 nm laser process. The lifetime reduction was below 20 % at the via hole. The superior performance of the 532 nm laser process is correlated to a smoother via surface and enhanced via sidewall oxidation. Finally, large area SHJ-MWT solar cells based on the optimized via formation process are processed and analyzed. (C) 2015 The Authors. Published by Elsevier Ltd.

Fixed charge and interface defect densities are the critical material parameters for silicon surface passivation. These parameters are measured with high spatial resolution on 150 mm wafers using a novel method called -BiasMDP'. In this method the effective carrier lifetime is determined by means of microwave detected photoconductivity while a bias voltage is applied to an electrode on top of the passivation layer. The measured carrier lifetime strongly reduces when the external bias voltage compensates the electric field of the fixed charges. Based on this minimum lifetime fixed charge and interface defect densities are calculated. The sensitivity of BiasMDP is demonstrated on a silicon wafer with Al2O3 passivation, which is processed with thickness gradient. A continuous shift of flat band voltage is measured across the wafer. This shift linearly correlates with oxide thickness as predicted by theory. Furthermore, an Al2O3 passivation stack with local HfO2 interface layer is characterized. The 2D map of fixed charge density shows a variation between 0.5 x 10(12) cm(-2) and 4.0 x 10(12) cm(-2) of negative polarity. This inhomogeneity is attributed to the local presence or not presence of the HfO2 interface layer. Finally, a wafer with local surface damage and inhomogeneous carrier lifetime distribution is measured with BiasMDP. The measurement shows that the inhomogeneity is caused by degradation of chemical passivaton. BiasMDP is shown to be a powerful method for detecting inhomogeneities of passivation layers. (C) 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Mueller matrix spectroscopic ellipsometry becomes increasingly important for determining structural parameters of periodic line gratings. Because of the anisotropic character of gratings, the measured Mueller matrix elements are highly azimuthal angle dependent. Measurement results are interpreted by basic principles of diffraction on gratings. The spectral and azimuthal angle dependent intensity changes are correlated to so-called Rayleigh singularities, i.e., wavelengths where the number of diffraction orders changes. The positions of the Rayleigh singularities are calculated analytically and overlapped with measured spectra of two different types of photomasks with transparent and reflecting substrates. For both types of gratings, the Rayleigh singularities reproduce the contours of the spectra. Increasing grating periods result in a shift of these contours to longer wavelengths. Characteristic differences between the two photomasks are explained by the influence of the transmission orders, which are determined by the substrate transparency.

A characterization method is presented, which determines fixed charge and interface defect densities in passivation layers. This method bases on a bias voltage applied to an electrode on top of the passivation layer. During a voltage sweep, the effective carrier lifetime is measured by means of microwave detected photoconductivity. When the external voltage compensates the electric field of the fixed charges, the lifetime drops to a minimum value. This minimum value correlates to the flat band voltage determined in reference impedance measurements. This correlation is measured on p-type silicon passivated by Al2O3 and Al2O3/HfO2 stacks with different fixed charge densities and layer thicknesses. Negative fixed charges with densities of 3.8 × 1012 cm−2 and 0.7 × 1012 cm−2 are determined for Al2O3 layers without and with an ultra-thin HfO2 interface, respectively. The voltage and illumination dependencies of the effective carrier lifetime are simulated with Shockley Read Hall surface recombination at continuous defects with parabolic capture cross section distributions for electrons and holes. The best match with the measured data is achieved with a very low interface defect density of 1 × 1010 eV−1 cm−2 for the Al2O3 sample with HfO2 interface.

This value is achieved due to a very low interface trap density of below 1010 eV–1 cm–2 and a fixed charge density of (2–3) × 1012 cm–2. In contrast, plasma ALD‐grown Al2O3 layers only reach carrier lifetimes of about 1 ms. This is mainly caused by a more than 10 times higher density of interface traps, and thus, inferior chemical passivation. The strong influence of the deposition parameters is explained by the limitation of hydrogen transport in Al2O3 during low‐thermal budget annealing. (© 2015 WILEY‐VCH Verlag GmbH &Co. KGaA, Weinheim)

Al2O3-TiO2 nanolaminates are very attractive candidates for future conductive passivation layers because they are purely based on dielectric materials, which allow a simple integration in the state-of-the-art manufacturing process. In this study, Al2O3-TiO2 double and multilayers are grown by atomic layer deposition and systematically investigated. The nanolaminates feature good silicon surface passivation and moderate electrical conductivity. The best performance is found for a double-layer stack consisting of a 5 nm Al2O3 interface layer and a 15 nm TiO2 capping layer after postdeposition annealing in N2 or forming gas. With this stack, a surface recombination velocity of 15 cm/s and a contact resistance of 20 Ω·cm2 are achieved. In Al2O3-TiO2 nanolaminates, the electrical transport is strongly influenced by the interaction of TiO2 and Al2O3 during layer growth. Raman measurements reveal that high conductivity correlates with a phase transition of TiO2 from amorphous to anatase.

Al2O3 based nanolaminates provide new functionalities for silicon surface passivation layers in future high-efficient cell concepts. This work presents two different applications: (I) Symmetrical passivation layers for application on p-and n-type Si are realized with thin HfO2 and Al-doped SiO2 interface layers between Al2O3 and Si. These stacks have excellent chemical passivation but zero fixed charges. Consequently, no surface inversion layer is formed in n-type Si and low surface recombination velocities below 2 cm/s are achieved in low injection. (II) Conductive passivation layers are realized with Al2O3-TiO2 double and multilayers. The optimum material combination and post deposition treatment results in a current density of 0.5 mA/cm(2) at 10 mV with a surface recombination velocity of about 15 cm/s. This conductivity is more than eight orders of magnitude higher than for pure Al2O3.

A controlled field-effect passivation by a well-defined density of fixed charges is crucial for modern solar cell surface passivation schemes. Al2O3 nanolayers grown by atomic layer deposition contain negative fixed charges. Electrical measurements on slant-etched layers reveal that these charges are located within a 1 nm distance to the interface with the Si substrate. When inserting additional interface layers, the fixed charge density can be continuously adjusted from 3.5 × 10(12) cm(-2) (negative polarity) to 0.0 and up to 4.0 × 10(12) cm(-2) (positive polarity). A HfO2 interface layer of one or more monolayers reduces the negative fixed charges in Al2O3 to zero. The role of HfO2 is described as an inert spacer controlling the distance between Al2O3 and the Si substrate. It is suggested that this spacer alters the nonstoichiometric initial Al2O3 growth regime, which is responsible for the charge formation. On the basis of this charge-free HfO2/Al2O3 stack, negative or positive fixed charges can be formed by introducing additional thin Al2O3 or SiO2 layers between the Si substrate and this HfO2/Al2O3 capping layer. All stacks provide very good passivation of the silicon surface. The measured effective carrier lifetimes are between 1 and 30 ms. This charge control in Al2O3 nanolayers allows the construction of zero-fixed-charge passivation layers as well as layers with tailored fixed charge densities for future solar cell concepts and other field-effect based devices.

Critical dimension and line edge roughness on photomask arrays are determined with Mueller matrix spectroscopic ellipsometry. Arrays with large sinusoidal perturbations are measured for different azimuth angels and compared with simulations based on rigorous coupled wave analysis. Experiment and simulation show that line edge roughness leads to characteristic changes in the different Mueller matrix elements. The influence of line edge roughness is interpreted as an increase of isotropic character of the sample. The changes in the Mueller matrix elements are very similar when the arrays are statistically perturbed with rms roughness values in the nanometer range suggesting that the results on the sinusoidal test structures are also relevant for “real” mask errors. Critical dimension errors and line edge roughness have similar impact on the SE MM measurement. To distinguish between both deviations, a strategy based on the calculation of sensitivities and correlation coefficients for all Mueller matrix elements is shown. The Mueller matrix elements M13/M31 and M34/M43 are the most suitable elements due to their high sensitivities to critical dimension errors and line edge roughness and, at the same time, to a low correlation coefficient between both influences. From the simulated sensitivities, it is estimated that the measurement accuracy has to be in the order of 0.01 and 0.001 for the detection of 1 nm critical dimension error and 1 nm line edge roughness, respectively.

Nanolaminates comprising of TiO2 or HfO2 sublayers within an Al2O3 matrix are grown with atomic layer deposition. These nanolaminates provide an improved silicon surface passivation compared to conventional Al2O3 films. The physical properties of the nanolaminates can be described with a dynamic growth model that considers initial and steady-state growth rates for the involved metal oxides. This model links the cycle ratios of the different atomic layer deposition precursors to the thickness and the material concentrations of the nanolaminate, which are determined by means of spectroscopic ellipsometry. Effective carrier lifetime measurements show that Al2O3-TiO2 nanolaminates achieve values of up to 6.0 ms at a TiO2 concentration of 0.2%. In Al2O3-HfO2 nanolaminates, a maximum effective carrier lifetime of 5.5 ms is reached at 7% HfO2. Electrical measurements show that the TiO2 incorporation causes strong hysteresis effects, which are linked to the trapping of negative charges and result in an enhanced field effect passivation. For the Al2O3-HfO2 nanolaminates, the capacitance data clearly show a very low density of interface traps (below 5·1010 eV−1·cm−2) and a reduction of the fixed charge density with increasing HfO2 concentration. Due to the low number of recombination centers near the surface, the reduced field effect passivation only had a minor impact on the effective carrier lifetime.

On n-type silicon, negatively charged surface passivation layers create a near surface recombination channel, which could significantly reduce the effective carrier lifetime at low injection levels (Δn < 1014 cm−3). This effect is described by Shockley Read Hall recombination at homogeneously distributed defects in the silicon wafer. In the near surface region, fixed charges in the dielectric layer significantly change the carrier concentrations and the recombination rate of defects. Sentaurus device simulations show that the contribution of the near surface recombination to the effective carrier lifetime depends on the properties of the involved defects. The lifetime reduction is strongest when the involved defects have an energy level in the lower half of the band gap and a very high electron to hole capture cross section ratio. For the simulation, a very low defect density in the order of 108 cm−3 is assumed, which is a realistic value in highly pure float zone silicon. Quasi-steady state photoconductance measurements on n-type silicon with Al2O3 passivation are done and fitted with the recombination model. Very good correlation between simulation and experiment is achieved when the involved recombination centers have an electron to hole capture cross section ratio of 107 and an energy level of −0.2 eV w.r.t. the intrinsic level. The simulated defect properties are discussed in respect of transition metal and doping related defects reported in literature.

Al2O3 nanolayers, as currently used in the solar industry, provide excellent passivation over the entire injection level range for p-type Si. A high concentration of negative fixed charges guarantees excellent field effect passivation. However, on n-type Si, those fixed charges create an inversion layer at the surface, which opens an additional near-surface recombination channel. This results in a reduced minority carrier lifetime at low injection levels, which is the operation condition of solar cells. In this work, a passivation layer stack is presented, showing excellent performance within the complete injection level range symmetrically for both p- and n-type Si. This layer stack consists of 20nm Al2O3 combined with a thin HfO2 or SiO2 interface. Electrical measurements show a reduction of the fixed charge density to virtually ‘zero’, when the interface layer is adjusted accordingly. This affects the field effect passivation and leads to a decrease of the near surface recombination in n-type Si. The introduction of five cycles of atomic layer deposited HfO2 improves the effective carrier lifetime in passivated n-type Si from 6 to 15ms at an injection level of 5×1012cm−3. The same passivation stack hardly influences the lifetime in p-type Si. The effective surface recombination velocity reaches values below 1cm/s for both types of substrate doping. In case of the introduction of a 1nm thick SiO2 interface, which also results in ‘zero’ fixed charges, the surface recombination velocity reaches a value of up to 200cm/s due to enhanced surface recombination. The comparison of both passivation stacks shows that the introduction of an ultrathin HfO2 interface is the most promising approach to achieve a symmetrical passivation layer for p- and n-type Si.