High throughput laser-scribing processes for industrial production of flexible CIGS thin-film solar modules Andreas Burn a , Christian Heger a , Stephan Buecheler b , Lukas Greuter b , Patrick Reinhard c , Roger Ziltener c , Lukas Krainer d , Gabriel Spuehler d , Valerio Romano a a Bern University of Applied Sciences, Applied Laser-, Photonics- and Surface Technologies ALPS, Pestalozzistrasse 20, 3400 Burgdorf, Switzerland; b Laboratory for Thin Films and Photovoltaics, Empa, Swiss Federal Laboratories for Materials Science and Technology, Ueberlandstrasse 129, 8600 Duebendorf, Switzerland; c Flisom AG, Ueberlandstrasse 129, 8600 Duebendorf, Switzerland; d Onefive GmbH, Althardstrasse 70, 8105 Regensdorf, Switzerland Author e-mail address: burn.andreas@gmail.com Abstract: Robust high-throughput laser scribing processes for monolithic interconnection of Cu(In,Ga)Se 2 absorber based thin-film solar cells were developed, validated and assessed for industrial roll-to-roll production of photovoltaic modules. Here we present results of the FP7-project APPOLO. OCIS codes: (140.0140) Lasers and laser optics; (350.3390) Laser material processing; (310.0310) Thin films 1. Introduction Absorber layers based on Cu(In,Ga)Se 2 are attractive for solar cell production because of their potentially high maximum efficiency and tunable band gap. This high potential resulted in best of class efficiencies above 20 percent for small single cells, e.g. ZSW 22.6%[1]. In recent years tremendous efforts have been made to up-scale the thin-film deposition processes to industrial production capacities while conserving as much as possible of the high conversion efficiency. CIGS thin-film technology is compatible with roll-to-roll production thanks to available low- temperature deposition processes[2]. An important step when moving from single solar cells to modules is the electrical series interconnection of the cell strips. Laser scribing is key enabling technology to the monolithic interconnection of cells into modules on flexible substrate because traditional techniques such as needle scribing, etching or shindling are either not applicable or not desirable. Our research consortium previously investigated several different process regimes including lift-off, induced lift-off, and direct ablation for P1 (electrical separation of the back contact), P2 (electrical connection between back- and front contact), and P3 (electrical separation of the front contact). Promising scribing processes were validated in mini modules produced on glass substrate. Best performance demonstrated was 16.6 percent conversion efficiency (aperture area) for a purely laser-structured grid-less 8-cell mini module[3] (certified by Fraunhofer ISE). 3. Scope of the project In the frame of the FP7 project APPOLO (Hub of Application Laboratories for Equipment Assessment in Laser Based Manufacturing) we aim to transfer our optimized scribing processes to industrial scale and to flexible substrates. The consortium involving all key partners along the value chain followed a three stage assessment process: i) optimization and validation of the laser source, ii) optimization and validation of scribing processes at industrially relevant throughput, and iii) assessment of laser equipment and scribing processes in industrial module production. The industrial scenario used was as follows: in-line scribing integrated in a roll-to-roll production line with a production rate of 0.5 to 1 m/min and a roll width of 0.5-1 m. Scribing velocities requested in this scenario are 1-2 m/s. The modules were produced in substrate configuration e.g. with the following full-stack order: substrate/Mo/Cu(In,Ga)Se2/CdS/ZnO/ZnO:Al (substrate/back contact/absorber/buffer/transparent front contact). 4. Scribing processes The well-established P1 lift-off process is a low overlap process with illumination of the molybdenum back contact layer through the transparent substrate [4]. Scaling of the process throughput to several meters per second by increasing the laser pulse repetition frequency was demonstrated and no degradation was observed. Typical scribing parameters for P1 and subsequent scribing processes are given in Table 1. The classical P2 process, where the absorber layer is directly ablated with high pulse-to-pulse overlap, is the most critical for up-scaling. Increasing the laser pulse repetition frequency leads to heat accumulation in the absorber layer, an unwanted effect that limits the scribing velocity to <100 mm/s on the material used in this study. Our solution to the problem is energy scaling of the process: a high energy laser pulse is focused to a narrow line with a top hat profile in direction of scribing. In this way, laser fluence and pulse to pulse overlap and pulse repetition frequency can be kept constant. The scribing velocity scales approximately proportional to the aspect This document is the accepted manuscript version of the following article: Burn, A., Heger, C., Buecheler, S., Greuter, L., Reinhard, P., Ziltener, R., … Romano, V. (2017). High throughput laser-scribing processes for industrial production of flexible CIGS thin-film solar modules. In Conference on lasers and electro-optics (p. ATu1C.4 (2 pp.). San Jose, California, United States: OSA Publishing. http://doi.org/10.1364/CLEO_AT.2017.ATu1C.4