Search

The silicon solar cells achieved relatively low prices in the last years and to introduce a new structure in the PV industry, the amount of silicon per watt has to be reduced, requiring a cost-effective manufacturing process. The use of n-type solar grade silicon has the advantages of presenting higher minority carrier lifetime than p-type one and the absence of the boron-oxygen defects. The aim of this paper is to present the development of 100 μm thick n+np+ silicon solar cells with a selective p+ rear emitter formed by boron deposited by spin-on and an Al/Ag grid deposited by screen-printing. The firing temperature of Ag/Al (rear face) e Ag (front face) was optimized and the temperature of 840 °C produced the devices with higher efficiency. The solar cells presented efficiencies of 16%, achieving a low silicon consumption of 1.6 g/W, 40% lower than thick p-type devices produced by the same process.

Resumo O objetivo deste trabalho foi analisar a passivação de superfícies com SiO2 em células solares bifaciais fabricadas em lâminas finas bem como avaliar a influência do tipo de silício monocristalino utilizado para fabricar as células solares. A estrutura implementada foi a n+pp+, com fósforo na face frontal e boro e alumínio na face posterior, formando uma região seletiva de campo retrodifusor. Foram usadas lâminas de silício Cz (Czochralski) e FZ (float zone) com 145 μm de espessura e implementou-se uma etapa térmica para crescer uma camada de SiO2 para passivar as superfícies. O óxido passivou as superfícies, proporcionando um aumento de 0,5% absolutos na eficiência das células solares fabricadas em silício FZ. As células solares produzidas em Si-Cz não obtiveram o mesmo aumento porque apresentaram uma elevada resistência em série. A célula solar fina mais eficiente foi fabricada com silício FZ e com superfícies passivadas, atingindo eficiência de 14,5% e de 11,1%, para iluminação pela face n+ e p+, respectivamente. As células em Si-Cz, mesmo com menor tempo de vida inicial dos portadores de carga minoritários, atingiram eficiências um pouco menores que estes valores.

The diffusion of phosphorus and aluminum in a single-step thermal process may reduce the production cost of multicrystalline silicon (mc-Si) solar cells. The goal of this paper is to analyze the influence of phosphorus diffusion parameters on the electrical characteristics of p-type mc-Si solar cells processed with a single-step diffusion of phosphorus and aluminum. To avoid the wafer bowing during aluminum paste firing, the aluminum was deposited by e-beam and co-diffused in the same thermal step that the phosphorus diffusion. First, the aluminum diffusion was performed and, then, the POCl3 was introduced into the quartz tube. The steps of the phosphorus diffusion were experimentally optimized. The best result was found with the temperature of 875 oC. The POCl3 concentration of 0.15 % and the optimized phosphorus diffusion parameters lead to the efficiency of 14.1%. The reduction of the oxygen flow during ramp temperature and aluminum diffusion improved the efficiency.

RESUMO No custo de produção de uma célula solar, a lâmina de silício representa da ordem de 50 % do valor final do dispositivo. Para reduzir os custos, vem sendo proposta a diminuição das espessuras atuais de 180-200 µm para 100-120 µm. Embora o silício tipo p seja o mais usado na fabricação de células solares, este material é suscetível à degradação devido à interação boro-oxigênio. Os dispositivos fabricados em lâminas de Si tipo n, dopadas com fósforo não apresentam tal degradação e têm potencial de obtenção de altas eficiências devido ao maior tempo de vida dos portadores de carga minoritários. O objetivo deste trabalho é apresentar o desenvolvimento de células solares n+np+ fabricadas sobre lâminas de silício Cz (Czochralski) com 100 μm de espessura. Foram fabricados dispositivos finos com a estrutura n+np+ e comparados com os dispositivos n+pp+ fabricados em lâminas convencionais. A eficiência média alcançada foi de (14,6 ± 0,4) % e o dispositivo mais eficiente atingiu 15,3 %. O maior valor de eficiência ficou abaixo do obtido com estruturas n+pp+ de 200 μm de espessura, quando atingiu-se uma eficiência de 16,2 %. O principal parâmetro que reduziu a eficiência das células finas foi o fator de forma, que alcançou um valor máximo de 0,748. Embora a eficiência seja menor com as lâminas finas, obteve-se uma baixa relação massa/potência, de 1,5 g/W, valor que é aproximadamente 53 % inferior ao obtido quando foram usadas lâminas espessas tipo p. ABSTRACT In the solar cell production cost, the silicon wafer represents roughly 50 % of the final value of the device. To lower the costs, the reduction of the current wafer thickness from 180-200 µm to 100-120 µm has been proposed. Although the p-type silicon is used to manufacture solar cells, it is susceptible to degradation due to boron-oxygen interaction. The devices manufactured in n-type silicon wafers doped with phosphorus do not present such degradation and has the potential of achieving higher efficiencies due to the larger minority charge carrier lifetime. The aim of this work is to present the development of n+np+ solar cells fabricated in Cz silicon wafers 100 μm thick. Thin devices were manufactured with the n+np+ structure and compared to n+pp+ devices manufactured in standard thick wafers. The average efficiency of (14.6 ± 0.4) % was achieved and the most efficient device reached 15.3%. The efficiency was lower than obtained with n+pp+ thick devices, which presented 16.2 %. The main parameter that reduced the efficiency of thin cells was the fill factor, which reached a maximum value of 0.748. Although the efficiency reached by thin n-type devices was worse than thick p-type ones, the former achieved a low mass/power ratio of 1.5 g/W, which is approximately 53 % lower than that obtained from thick p-type devices.

N-type silicon solar cells are being investigated due to the potential to obtain high efficiency. In this solar cell, the front p+ emitter is performed with boron doping and the Al or Al/Ag screen printing pastes don’t reach the development of Ag pastes used to form the front contact in n+pp+ solar cells. Moreover, the rear area covered by metal grid on the phosphorus back surface field may be optimized to increase the silicon solar cell efficiency and reduce the cost production. The goal of this paper is to present the analysis of PV modules and silicon solar cells developed on phosphorus-doped Czochralski wafers with different metal grid on the rear face. The solar cells were processed by using spin-on dopant to obtain the boron emitter and the back surface field was formed by phosphorus diffusion. The front metal grid was formed with Al/Ag paste and the rear area covered by the Ag metal paste was varied from 9% to 53%. Solar cells were manufactured, characterized and sorted taking into account the short-circuit current. Eight PV modules were fabricated using a standard process. Solar cells presented low fill factor. When the Ag/Sn/Cu ribbon was soldered on the front busbars of a solar cell, the fill factor rose from 0.70 to 0.75, causing an increase of 13.2% to 15.0% in the efficiency. Therefore, the fill factor of the PV modules was higher than that of solar cells. The reduction of the metal grid area on the rear face of solar cells did not affect the fill factor and solar cells with low rear metal coverage presented higher short-circuit current. The infrared thermography analysis was performed under a cloudless sky day with modules in the short-circuit condition. All modules presented hot spots and this result was not related to the different metallized rear area. The temperature difference between the solar cells of each photovoltaic module varied from 9 °C to 18 °C.

A redução no custo de fabricação de células solares, mantendo a eficiência de conversão de energia solar em elétrica, é uma das principais metas de pesquisa, na área de energia solar fotovoltaica. O objetivo desse trabalho está centrado no desenvolvimento e na análise de um processo de fabricação de células solares industriais de silício, com região de campo retrodifusor, formada pela deposição de pasta de Al, por serigrafia e difusão em forno de esteira. A queima das pastas de prata e a queima de prata-alumínio, usadas para formar a malha metálica frontal e posterior das células solares e a difusão de Al foram realizadas simultaneamente. A melhor célula foi fabricada com processo de difusão/queima a temperatura de 860ºC e velocidade de esteira de 150 cm/min, resultando em dispositivos de 12,8% de eficiência. A região de campo retrodifusor obtida não foi tão efetiva quanto à obtida por meio da deposição de alumínio em alto vácuo e por meio da difusão em forno convencional ou forno de esteira. Cost reduction of solar cells while maintaining high solar energy conversion efficiencies is one of the main research goals in photovoltaic solar energy. The aim of this work is centered on the development of a fabrication process of industrial silicon solar cells with a back surface field region obtained by Al deposited by screen-printing and diffused into silicon wafer in a belt furnace. Ag and Ag-Al pastes were fired to establish the metallic contacts simultaneously to the Al diffusion. The best cell was processed at a temperature of 860°C and belt speed of 150 cm/min, achieving an efficiency of 12.8%. The BSF region was not as effective as that obtained from Al deposited at high vacuum and diffused in standard tube furnaces or belt furnaces.

RESUMO A passivação das superfícies das células solares de silício é importante para obter alta eficiência, pois reduz a recombinação dos portadores de carga minoritários. O crescimento de SiO2 é a técnica mais efetiva para passivar lâminas de silício cristalino. O objetivo deste trabalho é apresentar o desenvolvimento e a análise da passivação na face posterior e frontal com diferentes espessuras de SiO2, crescido por oxidação seca. Células solares foram processadas com campo retrodifusor seletivo em lâminas de silício Czochralski, tipo p, grau solar. O emissor frontal foi produzido pela difusão de fósforo e o campo retrodifusor seletivo foi formado pela difusão de boro e alumínio, para possibilitar a passivação também na face posterior. Variou-se a temperatura (Toxi) e o tempo (toxi) de oxidação para formar camadas de SiO2 com diferentes espessuras. Com base em trabalhos anteriores, fixou-se toxi em 7 minutos e variou-se Toxi de 770 °C a 920 °C. A seguir, fixou-se Toxi em 800 °C e variou-se toxi de 2 a 90 minutos. As espessuras das camadas de SiO2 produzidas variaram entre 7 e 80 nm no emissor frontal e entre 1 e 14 nm na face posterior. Para avaliar a passivação, comparou-se a tensão de circuito aberto (Voc) e constatou-se que aumentou até toxi = 45 min e até Toxi = 860 °C. Com base nestes resultados, variou-se o tempo de oxidação para Toxi = 860 °C. A maior Voc, de 604 mV, foi obtida para: a) Toxi = 860 °C e toxi = 30 min e b) Toxi = 800 °C e toxi = 45 min, resultando em uma camada de SiO2 da ordem de 50 - 60 nm e de 10 - 14 nm na face dopada com fósforo e com boro, respectivamente. A maior eficiência alcançada foi de 16,0 %. ABSTRACT The silicon solar cell passivation is essential to achieve high efficiency, because reduces the minority charge carrier recombination. The SiO2 growth is the most effective technique to passivate crystalline silicon wafers. The goal of this paper is to present the development and analysis of the rear and front passivation with different thicknesses of SiO2, grown by dry oxidation. Solar cells were processed with selective back surface field (BSF) in solar grade p-type Czochralski-grown wafers. The front emitter was produced by phosphorus diffusion and the selective BSF was formed by boron and aluminum diffusion to allow the passivation also in the rear face. The oxidation temperature (Toxi) and time (toxi) were ranged to form SiO2 layer with different thicknesses. Based on previous studies, the Toxi was kept at 800 °C and the toxi was ranged from 2 minutes to 90 minutes. In the next step, the Toxi was ranged from 770 °C to 920 °C for the toxi = 7 minutes. The thicknesses of the SiO2 layer ranged from 7 nm to 80 nm in the front emitter and from 1 nm to 14 nm in the rear face. The open circuit voltage (Voc) was compared to evaluate the passivation and this parameter increased up to toxi = 45 minutes and up to Toxi = 860 °C. Based on these results, the oxidation time was ranged for Toxi = 860 °C. The higher Voc, of 604 mV, was obtained for: Toxi = 860 °C and toxi = 30 minutes and b) Toxi = 800 °C and toxi = 45 min, resulting in a SiO2 layer of around 50 - 60 nm and of 10 - 14 nm in the phosphorus and boron doped face, respectively. The highest achieved efficiency was of 16.0 %.

O Sol é fonte de energia renovável e o seu uso para produzir energia elétrica é uma das alternativas promissoras para enfrentar os desafios energéticos e ambientais do novo milênio. O silício é o segundo elemento mais abundante da Terra. Este material semicondutor é largamente usado na indústria de células solares e dispositivos de microeletrônica e permite a fabricação de dispositivos de alta durabilidade. O silício cristalino tipo n vem despertando o interesse mundial por apresentar menor degradação e maior tempo de vida dos portadores minoritários quando comparado com silício tipo p. O objetivo deste trabalho está centrado no desenvolvimento de um processo de fabricação industrial de células solares p+nn+, pseudoquadradas de 80 mm x 80 mm, em silício crescido por fusão zonal flutuante (Si-FZ) tipo n, com metalização por serigrafia. A região p+ foi produzida a partir de dopante com boro depositado por spin-on e difusão a alta temperatura em forno convencional com tubo de quartzo. A dopagem com boro foi otimizada considerando as características elétricas das células solares. O método consistiu em variar a temperatura de difusão de 900 ºC a 1020 ºC e tempo de difusão do processo de 10 min a 40 min. O processo de passivação de superfície foi avaliado utilizando SiO2 o que demonstrou não ser eficaz para o tipo de célula solar deste trabalho. Os melhores dispositivos foram fabricados com difusão de boro a 1000 ºC por 30 min, sem passivação de superfície, atingindo-se a eficiência de 14,6 %.

The aim of this work is to present the development and comparison of thin n+pp+ industrial bifacial silicon solar cells produced with the local screen-printed Al back surface field (BSF) to those with the selective BSF doped with aluminum-boron. To produce solar cells with selective BSF, the boron diffusion based on spin-on dopant was introduced in the process sequence. The thin SiO2 layer (10 nm) thermally grown did not produce good passivation on the rear face and wafers were contaminated during aluminum diffusion in the belt furnace. The implementation of selectively doped BSF improved the efficiency by reflecting minority charge carriers and the wafer contamination by belt furnace was compensated by boron diffusion. The bifacial solar cells with B-Al selective BSF achieved an efficiency of 13.7% / 8.9% (front / rear illumination) and presented lower sensitivity to the belt furnace processing and to the quality of the rear surface passivation.

A bifacial cell technology for Cz Si and evaporated contacts is presented. A p/sup +/nn/sup +/ structure on high resistivity material gives 17.7% for n/sup +/ side illumination and 15.2% for p/sup +/ side illumination. Cell performance is analyzed by fitting experimental measurements with PC1D. Analysis shows that p/sup +/ layer puts a limit to cell performance, mainly due to a high surface recombination velocity. The boron depleted zone near the surface also enhances recombination, but its effect can be reduced by performing a boron etch-back step in the process. Cells with boron etch-back give higher short-circuit current and a reduction of open-circuit voltage of around 10 mV. These results are consistent with the PC1D model.

The aim of this paper is to analyze the fabrication process of thin bifacial silicon solar cells concerning the order of diffusions to form p+ and n+ regions. The n+pp+ structure with the p+ selective region was implemented by using thin solar grade Czochralski silicon wafers. The whole rear face was doped with boron deposited by spin-on and thermally diffused and an Al metal grid was screen-printed and diffused. The phosphorus diffusion after the boron one produced the thinner n+ emitter and thinner dead layer, which allow the manufacturing of more efficient solar cells. Furthermore, the phosphorus diffusion at the end of processing promoted gettering, enhancing the minority charge carrier lifetime. Solar cells with the phosphorus diffusion after the boron one reached front and rear efficiencies of 14.0% and 10.4%, respectively, without any surface passivation.

Brazil is a country that receives a large amount of solar radiation. Moreover, high quality quartz mines has been found in its territory and the country has well-established metallurgical grade silicon industries. However, PV technology is not used in large scale and there are no active solar cell fabs. In this context, a pilot plant to develop industrial PV technology was carried out in a Brazilian university combining education, research and industrial development. Silicon solar cell processes based on low cost chemicals and gases as well as gettering mechanisms were developed and more than 12,000 solar cells were fabricated, achieving efficiencies up to 16%, and two hundred photovoltaic modules were produced and characterized.

The industrial process of silicon solar cells is mainly based on the aluminum back surface field (Al-BSF) performed by the conventional screen printing and firing in a belt furnace. The goal of this paper is to present the analysis of the influence of the amount of the Al paste on the electrical parameters and on the minority carrier diffusion length. The silicon solar cells were processed in p-type Czochralski solar grade wafers. The amount of Al paste deposited to form the BSF, denominated of Al paste surface density, was ranged from 2.8 mg/cm2 to 8.8 mg/cm2. The peak firing temperature for each amount of Al paste was optimized and the depth of the Al-BSF was estimated. The best results were found for the Al paste surface density of 3.5 mg/cm2 and the peak firing temperature of 840 °C, resulting in the efficiency of the solar cells of (15.0 ± 0.1) %. In this case, the depth of the Al-BSF was (5 ± 1) μm. Taking into account the peak firing temperature obtained for each Al paste surface density, we observed that the short-circuit current increased up to the average Al paste amount of 3.5 mg/cm2. On the other hand, the fill factor decreased with the increasing of the Al paste surface density. The open circuit voltage was slightly affected by the Al paste amount. The minority carrier lifetime rose from 30 μs to 120 μs after the phosphorus diffusion. A strong improvement in the minority carrier diffusion length was observed after the firing process and depends on the Al paste amount. The measured average values were 500 μm, 1280 μm and 780 μm for the Al paste surface density of about 2.8 mg/cm2, 3.5 mg/cm2 and 8.8 mg/cm2, respectively.

Up-converters use radiation normally transmitted by the solar cell and converts it to radiation, which can be converted to electric energy. Therefore, an up-converter should be placed on the rear side of a bifacial cell. This has the advantage that no impairment of the radiation impinging on the solar cell takes place. The number of photons that may be converted is limited by the absorption range of the up-converter and the efficiency of the up-conversion. High transmittance is required in the absorption range of the up-converter and the use of photoluminescence materials enhances the up-conversion phenomena. The photoluminescence material should absorb over a wide spectral range and emit in the absorption range of the up- converter. This paper presents the evaluation of the optical properties of the up-converter combined with PbS quantum dots and the reduction of the reflection losses in industrial bifacial silicon solar cells with up-converter/quantum dots incorporated. The up-converter materials are in the powder form, and the PbS quantum dots are available in the liquid form, so that an agent is needed, which at the same time should give a good optical coupling to the solar cell. Silicone gel with up- converter and quantum dots was laminated to the rear face of the bifacial cell by using an industrial technique based on frontsheet/EVA/backsheet. Combining the effects of up-converter and quantum dots, the reflection losses were reduced by 20%, demonstrating that the absorption and emission characteristics of the up-converter and quantum dots embedded in silicon can be tuned to the desired spectral region.