n-Type Rear Junction Solar Cells with Locally Contacted Al-Alloyed Emitter

Lade...
Vorschaubild
Dateien
DiplomaThesis_Raabe_flat.pdf
DiplomaThesis_Raabe_flat.pdfGröße: 4.25 MBDownloads: 963
Datum
2012
Autor:innen
Raabe, Johannes Nils
Herausgeber:innen
Kontakt
ISSN der Zeitschrift
Electronic ISSN
ISBN
Bibliografische Daten
Verlag
Schriftenreihe
Auflagebezeichnung
DOI (zitierfähiger Link)
ArXiv-ID
Internationale Patentnummer
Angaben zur Forschungsförderung
Projekt
Open Access-Veröffentlichung
Open Access Green
Sammlungen
Core Facility der Universität Konstanz
Gesperrt bis
Titel in einer weiteren Sprache
Forschungsvorhaben
Organisationseinheiten
Zeitschriftenheft
Publikationstyp
Masterarbeit/Diplomarbeit
Publikationsstatus
Published
Erschienen in
Zusammenfassung

This diploma thesis was focused on enhancing the rear side performance of the improved PhosTop solar cell concept by means of dielectric rear side passivation and reduction of the highly doped emitter area. Stack systems and passivation layers were applied on lowly doped n-type silicon bulk and highly doped aluminium p-type emitters in order to reduce the effective rear surface recombination velocity and hence improve the open-circuit voltage. Furthermore, using a dielectric rear passivation leads to an improved internal reflection on the rear side, which results in a short-circuit current density gain. Three different solar cell concepts were realized.

Besides the improved PhosTop solar cell, which is representing the reference, the Al-LARE (Aluminium - Locally Alloyed Rear Emitter) solar cell featuring a passivated n-type bulk and locally alloyed emitter is presented. Furthermore, the FALCON (Full Area Locally CONtacted emitter) solar cell is realized, which exhibits a full area alloyed and etched back emitter that is passivated and locally contacted.

n-type silicon substrates were passivated using different layers and stacks of which Al2O3/SiNA-SiNx, SiNA-SiNx and CT-SiNx passivation performed best after a firing step and featured effective lifetimes of up to 9.5 ms. In the contrary to highly doped n-type silicon, the SiO2/SiNA-SiNx passivation on n-type substrates showed a severe firing instability for temperatures above 800 celsius.
The characterization of the emitter formed by three different aluminium pastes revealed very low emitter saturation current densities in the range of 150 to 180 fA/cm2, showing further no influence on the set-firing peak temperature. The passivation of the etched back emitter was not found to be on a satisfactory level on samples that featured an etched back emitter, which is thicker than 1.5 µm, being below the passivation due to the field-effect of the non-etched back emitter. Compared to the fully alloyed and metallized rear side of the PhosTop solar cell, no
improvement of the rear could be made by using different passivation layers and stacks.

Al-LARE solar cells were simulated using PC2D, a novel two-dimensional simulation tool. The simulation predicted a possible short-circuit current density gain of 0.5 mA/cm2 for an emitter and contact width of 100 µm and emitter spacing of 200 to 300 µm. This is in good agreement with a followed emitter width and spacing variation that was carried out on 5x5 cm2 Al-LARE solar cells. The emitter width and spacing resulted in the conclusion that the highest values are obtained for a minimum of 100 µm emitter width and an emitter spacing of 300 - 400 µm (depending on the emitter width).

Large-area Al-LARE solar cells featuring an emitter width of 100 µm and an emitter spacing of 300 µm were further fabricated and analysed. The best performing Al-LARE solar cell that was passivated by SiO2/SiNA-SiNx, reached an efficiency of 17 %. Furthermore, a maximum short-circuit current density gain of 0.45 mA/cm2 compared to jointly fabricated PhosTop solar cells was found for a Al-LARE solar cell passivated by a Al2O3/SiNA-SiNx on the rear side. This solar cell concept was found to be basically limited by extremely high values of j02 in combination with an in some cases slightly increased series resistance due to contact formation problems at the rear side. Furthermore, the passivation quality of the Al2O3/SiNA-SiNx passivated rear was the only passivation able to compensate diffusion losses to the emitter and hence to sustain a comparable IQE plateau to the PhosTop solar cell.



Finally, FALCON solar cells were fabricated that feature an etched back 2 µm deep, screenprinted full area aluminium alloyed passivated emitter. Two different process sequences were carried out, allowing the rear side of one experiment to be passivated by a SiO2/SiNA-SiNx stack. An overall short-circuit current density gain, similar to the Al-LARE solar cells, of 0.5 mA/cm2 was found for the best performing SiO2/SiNA-SiNx FALCON solar cell. This is only half of the short-circuit current density gain that was expected from the simulation. This is probably caused by a discrepancy of the assumed reflection difference between the unpassivated and passivated rear side for the simulation and the difference for real solar cells. Furthermore, a strong discrepancy of in some cases almost 20 mV was found between the simulated and actually measured Voc. This discrepancy can be attributed to a lower passivation quality and hence the rear SRV in the fabricated experiments compared to the simulation and increased j02.

The best performing FALCON solar cell was achieved by passivating the rear using a SiO2/SiNA-SiNx stack that resulted in an efficiency of 18.9 %. Especially for low performing FALCON solar cells, a high reduction in FF and hence in efficiency was found to be due to an increased series resistance reaching approx. 1 Ohm cm2 and in some cases, an extremely high j02. j02 in combination with a not improved or even increased j01 compared to the non-passivated rear of the PhosTop solar cell, resulted in a moderate to strongly reduced Voc, as well.

The main advantage of the FALCON compared to the Al-LARE solar cell is the full area etched-back and passivated emitter that leads to a constant plateau in the IQE in the visible light range and hence allows a higher jsc compared to the decreased plateau of the Al-LARE solar cell. Furthermore, this full area emitter can lead to a much lower j02 that is found for FALCON solar cells compared to the Al-LARE solar cells.

In conclusion, since Al-LARE solar cells are mainly limited due to an extremely high j02, this diploma thesis suggests that unless improvements can be made, the increased fabrication effort is not justified, since a maximum obtained efficiency of 17 % is much lower than the 19.4 % of the improved PhosTop solar cell, while the latter is much easier to fabricate.

The FALCON solar cell concept has a higher potential, since it is mainly limited due to process parameters such as unfilled line contacts on the rear that result in an increased series resistance.

Furthermore, the passivation on the etched-back emitter needs to be further increased. For an industrial implementation of this solar cell concept the needed processing steps for fabrication have to be reduced.

Zusammenfassung in einer weiteren Sprache

Ziel dieser Diplomarbeit war die Verbesserung der Rückseite der PhosTop Solarzelle. Um dies zu erreichen, wurde die vollflächige Metallisierung durch eine Passivierungsschicht ersetzt, welche aus einer oder mehreren dielektrischen Schichten besteht. Eine dielektrische Schicht auf der Rückseite hat zweierlei Vorteile. Zum einen kann sie zu einer Erhöhung der offenen Klemmspannung, aber auch zu einer Verbesserung des Kurzschlussstroms (nicht bei Solarzellen mit Emittern auf der Rückseite) aufgrund verbesserter Passivierung der Siliziumoberfläche führen. Zum anderen ermöglicht das Aufbringen von dielektrischen Schichten auf der Rückseite die interne Reflexion zu verbessern, was wiederum zu einem Kurzschlussstromgewinn führen kann.

In dieser Arbeit sind zwei unterschiedliche Solarzellenkonzepte realisiert worden. Beide Konzepte sind mit der Vorderseite der PhosTop Solarzelle versehen, welche ein selektives Front Surface Field sowie eine dielektrische Passivierung bestehend aus einem Siliziumdioxid/Siliziumnitrid Stapel aufweist. Das Al-LARE Solarzellenkonzept hat auf der Rückseite einen lokal legierten
Emitter, der sich nach vollflächiger Passivierung und lokaler Öffnung durch vollflächigen Druck einer Paste und nachfolgendem Feuern bildet. Zum anderen wurde das FALCON Solarzellenkonzept realisiert, welches aus einem vollflächigen, zurückgeätzten Emitter besteht, der passiviert und lokal kontaktiert wird.

Zunächst wurden die Passivierungseigenschaften unterschiedlicher Schichten und Stacksysteme auf negativ sowie stark positiv dotiertem Silizium untersucht. Dabei wurde im ersteren Fall festgestellt, dass die Passivierung mit Al2O3/SiNA-SiNx, SiNA-SiNx und CT-SiNx die besten Resultate liefert. Es wurden zwei unterschiedliche Methoden zur Abscheidung einer Siliziumnitridschicht genutzt, deren Schichten in dieser Arbeit durch SiNA-SiNx (remote PECVD) und CT-SiNx (direct PECVD) bezeichnet werden. Im zweiten Fall, der Passivierung von hochdotierten Emittern wurde keine Verbesserung gegenüber der Feldeffektpassivierung durch den Emitter selbst gefunden. Eine Erniedrigung der effektiven Oberflächen- Rekombinationsgeschwindigkeit und eine damit einhergehende verbesserte Passivierung wurde nur für einen sehr dünnen Emitter erreicht, bei dem der Feldeffekt schon sehr schwach war. Entscheidend aber ist, das die effektive Passivierung eines hochdotierten Emitters mittels dielektrischer Schichten im Vergleich zu einer unpassivierten, nicht zurückgeätzen Referenz nicht verbessert wurde.

Aufbauend auf den Resultaten aus diesen beiden vorbereitenden Experimenten wurden großflächige Al-LARE und FALCON Solarzellen hergestellt, welche jeweils mit parallel prozessierten PhosTop Solarzellen verglichen wurden.

Um die optimalen Abstände und Breiten für die Rückseitengeometrie von Al-LARE Solarzellen zu bestimmen, wurde zunächst eine Emitterbreiten- sowie Abstandsvariation auf kleinen 5x5 cm2 Solarzellen durchgeführt. Diese zeigte für eine Emitterbreite von mindestens 100 µm und einen Emitterabstand von 300 µm die höchsten Kurzschlussströme. Das experimentelle Ergebnis konnte durch zwei-dimensionale Simulationen bestätigt werden. In beiden Fällen wurde durch die erhöhte interne Reflexion ein Kurzschlussstromgewinn von etwa 0.5 mA/cm2 erreicht.

Die beste 15.6x15.6 cm2 Al-LARE Solarzelle erreichte einen maximalen Wirkungsgrad von 17% mit einer Rückseitenpassivierung bestehend aus SiO2/SiNA-SiNx. Die Hauptgründe für eine Verschlechterung des Wirkungsgrades gegenüber der PhosTop Solarzelle waren zum einen ein sehr großer Wert für j02, zum anderen eine leicht verschlechterte effektive Oberflächenrekombinationsgeschwindigkeit.

Zusammen mit einem leicht erhöhten Serienwiderstand kam es zu einer Verschlechterung des Wirkungsgrades aufgrund einer starken Reduktion des Füllfaktors und der offenen Klemmspannung.
Für das FALCON Solarzellenkonzept wurden bessere Ergebnisse erzielt. FALCON Solarzellen, welche auf der Rückseite mit SiO2/SiNA-SiNx passiviert waren, erreichten einen Wirkungsgrad von 18.9%, wobei ein Kurzschlussstromgewinn im Vergleich zur sehr guten PhosTop Referenz von 0.5 mA/cm2 erreicht wurde. Auch bei diesem Zellkonzept waren wesentliche Limitierungen des Wirkungsgrades durch einen erhöhten Wert für j02, einen erhöhten Serienwiderstand und eine nicht verbesserte Rückseitenpassivierung (effektiv gegenüber vollflächiger Metallisierung) verursacht. Durch die ersten beiden Anteile wurde der FF um mindestens 2% reduziert, wobei die gegenüber der PhosTop Solarzelle nicht verbesserte Rückseitenpassivierung für keine Steigerung der offenen Klemmspannung verantwortlich ist. Eine Steigerung der offenen Klemmspannung ist, wie eine durchgeführte Simulation zeigte, nur für niedrigere Oberflächenrekombinationsgeschwindigkeiten möglich.

Fachgebiet (DDC)
530 Physik
Schlagwörter
Aluminium-Emitter, n-type, rear junction, PhosTop, Al-lare, Falcon
Konferenz
Rezension
undefined / . - undefined, undefined
Zitieren
ISO 690RAABE, Johannes Nils, 2012. n-Type Rear Junction Solar Cells with Locally Contacted Al-Alloyed Emitter [Master thesis]
BibTex
@mastersthesis{Raabe2012nType-22918,
  year={2012},
  title={n-Type Rear Junction Solar Cells with Locally Contacted Al-Alloyed Emitter},
  author={Raabe, Johannes Nils}
}
RDF
<rdf:RDF
    xmlns:dcterms="http://purl.org/dc/terms/"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:bibo="http://purl.org/ontology/bibo/"
    xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:void="http://rdfs.org/ns/void#"
    xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > 
  <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/22918">
    <dcterms:title>n-Type Rear Junction Solar Cells with Locally Contacted Al-Alloyed Emitter</dcterms:title>
    <dcterms:rights rdf:resource="https://rightsstatements.org/page/InC/1.0/"/>
    <dcterms:abstract xml:lang="eng">This diploma thesis was focused on enhancing the rear side performance of the improved PhosTop solar cell concept by means of dielectric rear side passivation and reduction of the highly doped emitter area. Stack systems and passivation layers were applied on lowly doped n-type silicon bulk and highly doped aluminium p-type emitters in order to reduce the effective rear surface recombination velocity and hence improve the open-circuit voltage. Furthermore, using a dielectric rear passivation leads to an improved internal reflection on the rear side, which results in a short-circuit current density gain. Three different solar cell concepts were realized.&lt;br /&gt;&lt;br /&gt;Besides the improved PhosTop solar cell, which is representing the reference, the Al-LARE (Aluminium - Locally Alloyed Rear Emitter) solar cell featuring a passivated n-type bulk and locally alloyed emitter is presented. Furthermore, the FALCON (Full Area Locally CONtacted emitter) solar cell is realized, which exhibits a full area alloyed and etched back emitter that is passivated and locally contacted.&lt;br /&gt;&lt;br /&gt;n-type silicon substrates were passivated using different layers and stacks of which Al&lt;sub&gt;2&lt;/sub&gt;O&lt;sub&gt;3&lt;/sub&gt;/SiNA-SiN&lt;sub&gt;x&lt;/sub&gt;, SiNA-SiN&lt;sub&gt;x&lt;/sub&gt; and CT-SiN&lt;sub&gt;x&lt;/sub&gt; passivation performed best after a  firing step and featured effective lifetimes of up to 9.5 ms. In the contrary to highly doped n-type silicon, the SiO&lt;sub&gt;2&lt;/sub&gt;/SiNA-SiN&lt;sub&gt;x&lt;/sub&gt; passivation on n-type substrates showed a severe firing instability for temperatures above 800 celsius.&lt;br /&gt;The characterization of the emitter formed by three different aluminium pastes revealed very low emitter saturation current densities in the range of 150 to 180 fA/cm&lt;sup&gt;2&lt;/sup&gt;, showing further no influence on the set-firing peak temperature. The passivation of the etched back emitter was not found to be on a satisfactory level on samples that featured an etched back emitter, which is thicker than 1.5 µm, being below the passivation due to the  field-effect of the non-etched back emitter. Compared to the fully alloyed and metallized rear side of the PhosTop solar cell, no&lt;br /&gt;improvement of the rear could be made by using different passivation layers and stacks.&lt;br /&gt;&lt;br /&gt;Al-LARE solar cells were simulated using PC2D, a novel two-dimensional simulation tool. The simulation predicted a possible short-circuit current density gain of 0.5 mA/cm&lt;sup&gt;2&lt;/sup&gt; for an emitter and contact width of 100 µm and emitter spacing of 200 to 300 µm. This is in good agreement with a followed emitter width and spacing variation that was carried out on 5x5 cm&lt;sup&gt;2&lt;/sup&gt; Al-LARE solar cells. The emitter width and spacing resulted in the conclusion that the highest values are obtained for a minimum of 100 µm emitter width and an emitter spacing of 300 - 400 µm (depending on the emitter width).&lt;br /&gt;&lt;br /&gt;Large-area Al-LARE solar cells featuring an emitter width of 100 µm and an emitter spacing of 300 µm were further fabricated and analysed. The best performing Al-LARE solar cell that was passivated by SiO&lt;sub&gt;2&lt;/sub&gt;/SiNA-SiN&lt;sub&gt;x&lt;/sub&gt;, reached an efficiency of 17 %. Furthermore, a maximum short-circuit current density gain of 0.45 mA/cm&lt;sup&gt;2&lt;/sup&gt; compared to jointly fabricated PhosTop solar cells was found for a Al-LARE solar cell passivated by a Al&lt;sub&gt;2&lt;/sub&gt;O&lt;sub&gt;3&lt;/sub&gt;/SiNA-SiN&lt;sub&gt;x&lt;/sub&gt; on the rear side. This solar cell concept was found to be basically limited by extremely high values of j&lt;sub&gt;02&lt;/sub&gt; in combination with an in some cases slightly increased series resistance due to contact formation problems at the rear side. Furthermore, the passivation quality of the Al&lt;sub&gt;2&lt;/sub&gt;O&lt;sub&gt;3&lt;/sub&gt;/SiNA-SiN&lt;sub&gt;x&lt;/sub&gt; passivated rear was the only passivation able to compensate diffusion losses to the emitter and hence to sustain a comparable IQE plateau to the PhosTop solar cell.&lt;br /&gt;&lt;br /&gt;&lt;br /&gt;&lt;br /&gt;Finally, FALCON solar cells were fabricated that feature an etched back 2 µm deep, screenprinted full area aluminium alloyed passivated emitter. Two different process sequences were carried out, allowing the rear side of one experiment to be passivated by a SiO&lt;sub&gt;2&lt;/sub&gt;/SiNA-SiN&lt;sub&gt;x&lt;/sub&gt; stack. An overall short-circuit current density gain, similar to the Al-LARE solar cells, of 0.5 mA/cm&lt;sup&gt;2&lt;/sup&gt; was found for the best performing SiO&lt;sub&gt;2&lt;/sub&gt;/SiNA-SiN&lt;sub&gt;x&lt;/sub&gt; FALCON solar cell. This is only half of the short-circuit current density gain that was expected from the simulation. This is probably caused by a discrepancy of the assumed reflection difference between the unpassivated and passivated rear side for the simulation and the difference for real solar cells. Furthermore, a strong discrepancy of in some cases almost 20 mV was found between the simulated and actually measured V&lt;sub&gt;oc&lt;/sub&gt;. This discrepancy can be attributed to a lower passivation quality and hence the rear SRV in the fabricated experiments compared to the simulation and increased j&lt;sub&gt;02&lt;/sub&gt;.&lt;br /&gt;&lt;br /&gt;The best performing FALCON solar cell was achieved by passivating the rear using a SiO&lt;sub&gt;2&lt;/sub&gt;/SiNA-SiN&lt;sub&gt;x&lt;/sub&gt; stack that resulted in an efficiency of 18.9 %. Especially for low performing FALCON solar cells, a high reduction in FF and hence in efficiency was found to be due to an increased series resistance reaching approx. 1 Ohm cm&lt;sup&gt;2&lt;/sup&gt; and in some cases, an extremely high j&lt;sub&gt;02&lt;/sub&gt;. j&lt;sub&gt;02&lt;/sub&gt; in combination with a not improved or even increased j&lt;sub&gt;01&lt;/sub&gt; compared to the non-passivated rear of the PhosTop solar cell, resulted in a moderate to strongly reduced V&lt;sub&gt;oc&lt;/sub&gt;, as well.&lt;br /&gt;&lt;br /&gt;The main advantage of the FALCON compared to the Al-LARE solar cell is the full area etched-back and passivated emitter that leads to a constant plateau in the IQE in the visible light range and hence allows a higher j&lt;sub&gt;sc&lt;/sub&gt; compared to the decreased plateau of the Al-LARE solar cell. Furthermore, this full area emitter can lead to a much lower j&lt;sub&gt;02&lt;/sub&gt; that is found for FALCON solar cells compared to the Al-LARE solar cells.&lt;br /&gt;&lt;br /&gt;In conclusion, since Al-LARE solar cells are mainly limited due to an extremely high j&lt;sub&gt;02&lt;/sub&gt;, this diploma thesis suggests that unless improvements can be made, the increased fabrication effort is not justified, since a maximum obtained efficiency of 17 % is much lower than the 19.4 % of the improved PhosTop solar cell, while the latter is much easier to fabricate.&lt;br /&gt;&lt;br /&gt;The FALCON solar cell concept has a higher potential, since it is mainly limited due to process parameters such as unfilled line contacts on the rear that result in an increased series resistance.&lt;br /&gt;&lt;br /&gt;Furthermore, the passivation on the etched-back emitter needs to be further increased. For an industrial implementation of this solar cell concept the needed processing steps for fabrication have to be reduced.</dcterms:abstract>
    <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/22918/2/DiplomaThesis_Raabe_flat.pdf"/>
    <dcterms:issued>2012</dcterms:issued>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/41"/>
    <dc:language>eng</dc:language>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2013-10-30T08:18:54Z</dcterms:available>
    <dc:creator>Raabe, Johannes Nils</dc:creator>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <dc:contributor>Raabe, Johannes Nils</dc:contributor>
    <dc:rights>terms-of-use</dc:rights>
    <bibo:uri rdf:resource="http://kops.uni-konstanz.de/handle/123456789/22918"/>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2013-10-30T08:18:54Z</dc:date>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/41"/>
    <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/22918/2/DiplomaThesis_Raabe_flat.pdf"/>
  </rdf:Description>
</rdf:RDF>
Interner Vermerk
xmlui.Submission.submit.DescribeStep.inputForms.label.kops_note_fromSubmitter
Kontakt
URL der Originalveröffentl.
Prüfdatum der URL
Prüfungsdatum der Dissertation
Finanzierungsart
Kommentar zur Publikation
Allianzlizenz
Corresponding Authors der Uni Konstanz vorhanden
Internationale Co-Autor:innen
Universitätsbibliographie
Begutachtet
Diese Publikation teilen