Filtern
Erscheinungsjahr
- 2011 (17) (entfernen)
Dokumenttyp
- Beitrag zu einem Tagungsband (7)
- Vortrag (4)
- Posterpräsentation (3)
- Buchkapitel (2)
- Zeitschriftenartikel (1)
Referierte Publikation
- nein (17) (entfernen)
Schlagworte
- Laser processing (4)
- Femtosecond laser ablation (3)
- Ablation (2)
- Damage threshold (2)
- Dielectrics (2)
- Doping (2)
- Incubation (2)
- Nanosecond laser ablation (2)
- Selective emitter (2)
- 532 nm wavelength (1)
Eingeladener Vortrag
- nein (4)
The selective emitter is a well-known technology for producing highly doped areas under the metallization grid to improve the solar cell performance. In this work, the influence of laser irradiation on phosphoric acid coated multicrystalline silicon PV-wafers on the wafer surface structure, the phosphorous depth distribution and the electrical contact resistance within the laser treated area as well as the electrical series resistance of laserprocessed solar cells was evaluated. Different laser processing settings were tested including pulsed and continuous wave (cw) laser sources (515 nm, 532 nm, 1064 nm wavelength). Complementary numerical simulations using the finite element method (FEM) were conducted to explain the impact of the laser parameters on the melting behavior (melt duration and geometry). It was found that the melt duration is a key parameter for a successful laser Doping process. Our simulations at a laser wavelengths of 515 nm reveal that low-repetition rate (<500 kHz) laser pulses of 300 ns duration generate a melt duration of ~0.35 µs, whereas upon scanning cw-laser radiation at 532 nm prolongates the melt duration by at least one order of magnitude. Experimentally, the widely used ns-laser pulses did not lead to satisfying laser irradiation results. In contrast, cw-laser radiation and scan velocities of less than 2 m/s led to suitable laser doping featuring low electrical resistances in the laser treated areas.
Towards an industrial laser doping process for the selective emitter using phosphoric acid as dopant
(2011)
Different laser supported approaches have already been realized, proving the great potential of laserdoped selective emitters (LDSE). However, it is challenging to establish a low-cost process by using pulsed laser tools. So far a single-step process only leads to satisfying results utilizing cw-lasers. In this paper we have examined a two-step process to produce laser-doped selective emitters on multicrystalline textured standard silicon photovoltaic wafers (90-Ω/sq-Emitter, SiN-antireflection coating (ARC)). The precise ARC removal by near-infrared fs-laser pulses (30 fs, 800 nm), and the doping of uncoated silicon wafers by ns-laser pulses (8 ns, 532 nm) were systematically investigated. In the fs-experiment, optimum conditions for ARC removal were identified. In the nsexperiments under suitable conditions (melting regime), the phosphorous concentration underneath the wafer surface was significantly increased and the sheet resistance was reduced by nearly a factor of two. Moreover, electrical measurements on fired metallization fingers deposited on the laser processed wafers showed low contact resistances. Hence, wafer conditioning with combined fs-laser- and ns-laser-processes are expected to be a promising technology for producing selective emitters.
Structuring of thin-film photovoltaic modules requires basic knowledge of the laser – thin-film interaction in order to adapt the accessible laser parameters, like wavelength, power, repetition rate and scribing speed whilst taking into account the specific material properties of the layer. We have studied the nanosecond laserablation behavior of corresponding layers (i) of silicon based thin-film solar cells with a-Si/µc-Si tandem absorber type and (ii) of back contact and absorber layer of CIGSe solar cells. The respective ablation threshold fluences were determined as integrative parameters describing the specific laser – material interaction. For the threshold determination we used two different methods and developed a new analytical approach taking into account scribing through the glass substrate as it is preferred for most structuring processes. This was done by analyzing the thin film ablation results by means of optical microscopy, profilometry, scanning electron microscopy (SEM). Moreover, we determined the incubation coefficient of the regarded material layers which allows us to predict quantitatively the influence of the spot overlap on the scribing threshold.
Manufacturing of CIGSe thin film solar modules involves typically one laser structuring step (P1) and two mechanical structuring steps (P2 and P3) for serial interconnection. In our approach, complete laser structuring is successfully demonstrated by application of short nanosecond laser pulses (<10 ns) with a single, visible wavelength of 532 nm. The P1 and the P3 trenches are scribed by induced and direct ablation, respectively. For the P2 scribe, the thermal input of the ns laser pulses is used to transform the CIGSe absorber layer locally into a highly conductive compound to provide proper electrical interconnection. These findings promise further simplification and flexibility to thin film solar cell production.
Ein wesentlicher Vorteil der Dünnschichtphotovoltaik gegenüber der waferbasierten Photovoltaik liegt in der monolithischen Serienverschaltung. Bei der Herstellung von Chalkopyrit(CIGSe) -Dünnschicht-Solarmodulen erfolgen dafür typischerweise ein Laser-Strukturierungsschritt (P1) und zwei mechanische Strukturierungsschritte (P2, P3). In diesem Beitrag wird gezeigt, dass die Strukturierung von CIGSe-Solarmodulen vollständig mit kurzen Laserpulsen (<10 ns Pulsdauer) und einer einzigen Wellenlänge (532 nm) möglich ist. Der P1- und P3-Schnitt erfolgen durch direkte induzierte Ablation. Für den P2-Schnitt wird gezielt der hohe Wärmeeintrag der ns-Laserpulse genutzt, um die CIGSe-
Absorberschicht lokal aufzuschmelzen und strukturell so zu verändern, dass eine elektrisch gut leitende Verbindung zwischen Front- und Rückkontakt entsteht.
Near-IR femtosecond (fs) (pulse duration = 150 fs, wavelength = 775 nm, repetition
rate 1 kHz) and VUV nanosecond (ns) (pulse duration = 20 ns, wavelength = 157 nm,
repetition rate 1 to 5 Hz) laser pulse ablation of single-crystalline TeO2 (c-TeO2 ) surfaces
was performed in air using the direct focusing technique. A multi-method
characterization using optical microscopy, atomic force microscopy and scanning
electron microscopy revealed the surface morphology of the ablated craters. This allowed
us at each irradiation site to characterize precisely the lateral and vertical dimensions of
the laser-ablated craters for different laser pulse energies and number of laser pulses per
spot. Based on the obtained information, we quantitatively determined the ablation
threshold fluence for the fs laser irradiation when different pulse numbers were applied to
the same spot using two independent extrapolation techniques. We found that in the case
of NIR fs laser pulse irradiation, the ablation threshold significantly depends on the
number of laser pulses applied to the same spot indicating that incubation effects play an
important role in this material. In the case of VUV ns laser pulses, the ablation rate is
significantly higher due to the high photon energy and the predominantly linear
absorption in the material. These results are discussed on the basis of recent models of the
interaction of laser pulses with dielectrics. In the second part of this chapter, we use time-
of-flight mass spectrometry (TOFMS) to analyze the elemental composition of the
ablation products generated upon laser irradiation of c-TeO2 with single fs- (pulse
duration ~200 fs, wavelength 398 nm) and ns-pulses (pulse duration 4 ns, wavelength
355 nm). Due to the three order of magnitude different peak intensities of the ns- and fs
laser pulses, significant differences were observed regarding the laser-induced species in
the plasma plume. Positive singly, doubly and triply charged Te ions (Te+, Te2+, Te3+) in
the form of many different isotopes were observed in case of both irradiations. In the case
of the ns-laser ablation, the TeO+ formation was negligible compared to the fs case and
there was no Te trimer (Te3+) formation observed. It was found that the amplitude of Te
ion signals strongly depends on the applied laser pulse energy. Singly charged oxygen
ions (O+) are always present as a byproduct in both kinds of laser ablation.