Search

In situ phosphorus (P)-doped polycrystalline silicon (poly-Si) films by low pressure chemical vapor deposition (LPCVD) were studied in this work for the fabrication of poly-Si passivating contacts. In situ doping was targeted for enabling the full potential of the high-throughput LPCVD technique, as it could allow leaner fabrication of industrial solar cells featuring poly-Si passivating contacts than the more common ex situ doping routes. By careful optimization of the deposition temperature and the flows of the carrier gas (H-2) and the dopant precursor (PH3), high doping in the poly-Si layers was achieved with active P concentrations up to 1.3.10(20) cm(-3) . While reduction in the deposition rate (r(dep)) and thus in the throughput is a known problem when growing in situ P-doped films by LPCVD, this reduction could be limited, and the resulting r(dep) was equal to 0.078 nm/s. The developed poly-Si films were characterized both structurally and in terms of their passivation potential in poly-Si contacts. The latter yielded recombination current densities down to 1.5 fA/cm(2) in passivated (J(0, p)) and 25.6 fA/cm(2) in screen-printing metallized (J(0, m)) regions on saw-damage removed (SDR) Cz-Si surfaces, accompanied by a contact resistivity (rho(c,m)) of 4.9 m Omega.cm(2). On textured Cz-Si surfaces, the corresponding values were J(0, p) = 3.5 fA/cm(2), J(0,m )= 56.7 fA/cm(2), and rho(c,m) = 1.8 m Omega.cm(2). Optical impact of the developed poly-Si films was also assessed and a short circuit density loss of 0.41 mA/cm(2) is predicted per each 100 nm of poly-Si applied at the rear side of solar cells. The authors would like to acknowledge Rajiv Sharma from KU Leuven for his help with the interfacial oxide development, Sukhvinder Singh and Patrick Choulat from Imec for their help with the contact resistivity measurements and sample fabrication, Thomas Nuytten and Stefanie Sergeant from Imec for the Raman spectroscopy measurements, Bastien Douhard and Mustafa Ayyad from Imec for SIMS measurements, Maxim Korytov, Laura Nelissen, and Patricia van Marcke from Imec for the TEM specimen preparation and measurements, and Janusz Bogdanowicz from Imec for his help with the analysis of the Hall measurement data. This work was supported by the European Union’s Horizon2020 Programme for research, technological development, and demonstration [grant number 857793]; and by the Kuwait for the Advancement of Sciences [grant number CN18-15EE-01].

Passivating contacts consisting of heavily doped polycrystalline silicon (poly-Si) and ultrathin interfacial silicon oxide (SiOx) films enable the fabrication of high-efficiency Si solar cells. The electrical properties and working mechanism of such poly-Si passivating contacts depend on the distribution of dopants at their interface with the underlying Si substrate of solar cells. Therefore, this distribution, particularly in the vicinity of pinholes in the SiOx film, is investigated in this work. Technology computer-aided design (TCAD) simulations were performed to study the diffusion of dopants, both phosphorus (P) and boron (B), from the poly-Si film into the Si substrate during the annealing process typically applied to poly-Si passivating contacts. The simulated 2D doping profiles indicate enhanced diffusion under pinholes, yielding deeper semicircular regions of increased doping compared to regions far removed from the pinholes. Such regions with locally enhanced doping were also experimentally demonstrated using high-resolution (5-10 nm/pixel) scanning spreading resistance microscopy (SSRM) for the first time. The SSRM measurements were performed on a variety of poly-Si passivating contacts, fabricated using different approaches by multiple research institutes, and the regions of doping enhancement were detected on samples where the presence of pinholes had been reported in the related literature. These findings can contribute to a better understanding, more accurate modeling, and optimization of poly-Si passivating contacts, which are increasingly being introduced in the mass production of Si solar cells. This work was supported by the European Union’s Horizon2020 Programme for research, technological development, and demonstration [Grant 857793] and by the Kuwait Foundation for the Advancement of Sciences [Grant CN18- 15EE-01].

Poly-Si/SiOx passivating contacts enable the manufacturing of highly-efficient Si solar cells, but their fabrication commonly relies on an extra high-temperature process such as dopant diffusion or thermal annealing for achieving excellent passivation and contacting properties. This extra process is eliminated in the fired passivating contact (FPC) approach used for simplified fabrication of poly-Si/SiOx passivating contacts. Instead, FPCs rely on the thermal budget of the fast/short and high-temperature firing process used for metallization of solar cells to achieve similar final properties. Despite this, compatibility of FPCs with industrially viable metallization techniques has not been demonstrated yet, which is studied in this work for fire-through Ag screen-printing and Ni/Ag plating. With screen-printing, low recombination current density (J(0)) down to 4.9 fA/cm(2), low contact resistivity between the Ag contacts and the FPC (rho(c,m)) down to 7.2 m Omega.cm(2), and Ohmic transport through the FPC including the SiOx film were achieved using wet-chemically grown SiOx. Nevertheless, J(0) of metallized regions (J(0,m)) exceeded 1000 fA/cm(2). Reducing J(0,m) was attempted by mitigating the blistering observed in FPCs, but J(0,m) remained high. With Ni/Ag plating, excellent surface passivation with J(0) down to 2.7 fA/cm(2) and very low J(0,m) < 50 fA/cm(2) were achieved, but no Ohmic contacts could be obtained. Integration of screen-printed FPCs in large-area n-TOPCon solar cells was also demonstrated, yielding average efficiencies of 18.4%, limited mainly by the high J(0,) (m) and series resistance of the FPCs. The results presented reveal the challenges for the industrialization of FPCs and provide valuable insights for tackling these. The authors would like to acknowledge Patrick Choulat from Imec and Rajiv Sharma from KU Leuven for the valuable discussions and their help with the sample fabrication, Thomas Nuytten and Stefanie Sergeant from Imec for the Raman spectroscopy measurements, and Bastien Douhard and Mustafa Ayyad from Imec for the SIMS measurements. This work was supported by the European Union’s Horizon 2020 Programme for research, technological development, and demonstration [grant number 857793]; and by the Kuwait Foundation for the Advancement of Sciences [grant number CN18-15EE-01].

The potential of in situ phosphorus (P)-doped polycrystalline silicon (poly-Si) films by low pressure chemical vapor deposition (LPCVD) was studied for the realization of poly-Si/SiO x passivating contacts. In situ doping of poly-Si, as an alternative to ex situ methods, could enable simpler fabrication of industrial solar cells featuring these passivating contacts. With this approach, recombination current densities down to 1.7 fA/cm2 and 3.5 fA/cm2 were achieved on saw-damage removed and textured Cz-Si surfaces, respectively. It was found that the use of thermal SiO x , high active doping in the poly-Si, and hydrogenation improve the passivation quality. In addition, while post-LPCVD annealing was also beneficial, dopant loss from poly-Si at high annealing thermal budgets was observed to be detrimental to the specific contact resistivity and passivation quality, thus making it crucial to mitigate such dopant losses.

The potential of passivating contacts incorporating in situ phosphorus (P)-doped polycrystalline silicon (poly-Si) films grown by low pressure chemical vapor deposition (LPCVD) is demonstrated in this work by integrating these layers at the rear side of large-area (241.3 cm(2)) bifacial n-type Tunnel Oxide Passivated Contact (nTOPCon) solar cells with diffused front emitter and screen-printed contacts. In situ doped poly-Si films are studied as their use could simplify the production of industrial n-TOPCon solar cells compared to the common approach relying on ex situ doping of intrinsic LPCVD poly-Si films. The developed poly-Si passivating contacts exhibited excellent characteristics with low recombination current densities in passivated and screen-printing metallized regions down to 2.3 fA/cm(2) and 65.8 fA/cm(2), respectively, and a low contact resistivity of 2.0 m Omega.cm(2). For reaching the best passivating contact characteristics and high solar cell efficiencies, a poly-Si film thickness of 150-200 nm was found to be optimal while a polished rear surface morphology was found to be beneficial. The best solar cell reached a certified power conversion efficiency of 23.01% along with a high open circuit voltage of 691.7 mV, enabled by the passivating contacts with the in situ doped poly-Si films. 1-cell glass-glass laminates were also fabricated with the developed solar cells, which showed no loss in their power output both upon 400 thermal cycles and after 1000 h of damp heat testing. Lastly, a roadmap is presented, indicating strategies to achieve efficiencies up to 25.5% with n-TOPCon solar cells incorporating the in situ P-doped LPCVD poly-Si films. The authors would like to acknowledge Sukhvinder Singh from Imec and Rajiv Sharma from KU Leuven for the valuable discussions and their help with the sample fabrication. This work was supported by the European Union’s Horizon2020 Programme for research, technological development, and demonstration [grant number 857793]; and by the Kuwait Foundation for the Advancement of Sciences [grant number CN18-15 EE-01].

We use temperature-dependent contact resistivity (ρc) measurements to systematically assess the dominant electron transport mechanism in a large set of poly-Si passivating contacts, fabricated by varying (i) the annealing temperature (Tann), (ii) the oxide thickness (tox), (iii) the oxidation method, and (iv) the surface morphology of the Si substrate. The results show that for silicon oxide thicknesses of 1.3–1.5 nm, the dominant transport mechanism changes from tunneling to drift-diffusion via pinholes in the SiOx layer for increasing Tann. This transition occurs for Tann in the range of 850°C-950 °C for a 1.5 nm thick thermal oxide, and 700°C-750 °C for a 1.3 nm thick wet-chemical oxide, which suggests that pinholes appear in wet-chemical oxides after exposure to lower thermal budgets compared to thermal oxides. For SiOx with tox = 2 nm, grown either thermally or by plasma-enhanced atomic layer deposition, carrier transport is pinhole-dominant for Tann = 1050 °C, whereas no electric current through the SiOx layer could be detected for lower Tann. Remarkably, the dominant transport mechanism is not affected by the substrate surface morphology, although lower values of ρc were measured on textured wafers compared to planar surfaces. Lifetime measurements suggest that the best carrier selectivity can be achieved by choosing Tann right above the transition range, but not too high, in order to induce pinhole dominant transport while preserving a good passivation quality.

Passivating contacts consisting of polycrystalline silicon (poly-Si) and thin silicon-oxide (SiOx) layers facilitate a significant reduction of recombination losses in silicon solar cells. Nevertheless, these gains come with short circuit current density (Jsc) losses due to parasitic absorption by the poly-Si. Even if the passivating contacts are employed at the rear side only, absorption, particularly due to free carriers (FCA) in the heavily doped poly-Si, may still lead to significant Jsc losses. In this work, these losses are characterized as a function of the poly-Si thickness (tpoly) by the analysis of front reflectance spectra in the infrared (IR). For this study, two sets of samples with different n-type full-area poly-Si passivating contacts at the rear are compared to references with a phosphorus(P)-diffused back surface field (BSF) instead. For the two sets, Jsc losses with respect to the references (ΔJsc) are 0.10 mA/cm2 and 0.42 mA/cm2 per 100 nm thick poly-Si, respectively. The difference between the two values is studied by Hall measurements and interpreted to be due to the over three times as high free carrier concentration (ND,act) in the poly-Si layers of the second set of samples as the first set. On the other hand, lifetime measurements showed an excellent passivation yielding an implied open circuit voltage (iVoc) up to 736 mV only for the samples with the more heavily doped poly-Si, whereas iVoc of 683 mV was measured for the first set, which indicates a trade-off between absorption losses and passivation quality.

We derive efficiency roadmaps for three industrial bifacial cell types, (1) pPERC, (2) rear-emitter nPERT, and (3) front-emitter nPERT, using an approach based on Quokka 2 numerical simulations. Using realistic incremental improvements based on data available in the literature, we show that the efficiency potential for bifacial pPERC and rear-emitter nPERT cells exceeds 24% while the efficiency potential for front-emitter nPERT cells is above 24.5%.We derive efficiency roadmaps for three industrial bifacial cell types, (1) pPERC, (2) rear-emitter nPERT, and (3) front-emitter nPERT, using an approach based on Quokka 2 numerical simulations. Using realistic incremental improvements based on data available in the literature, we show that the efficiency potential for bifacial pPERC and rear-emitter nPERT cells exceeds 24% while the efficiency potential for front-emitter nPERT cells is above 24.5%.