Large area plastic solar cell modules

https://doi.org/10.1016/j.mseb.2006.06.008Get rights and content

Abstract

Preliminary data on the fabrication of 0.1 m2 polymer solar cells are presented. The process employed screen-printing of an active layer onto an indium-tin-oxide (ITO) electrode pattern (50 Ω square−1) on a 200 μm polyethyleneterphthalate (PET) substrate. After the printing, vacuum coating of an optional layer of C60 and the final aluminium electrode was employed to complete the device. The active layer consisted of poly-1,4-(2-methoxy-5-ethylhexyloxy)phenylenevinylene (MEH-PPV). Chlorobenzene was used as solvent for the screen-printing process. The design of the solar cell module was chosen to employ both serial and parallel connection of individual solar cells. Thirteen individual solar cells with an active area of 7.2 cm2 were thus connected in series. The serial connection was chosen to reduce the current density for the large area employed. A step up in voltage is thus preferable to avoid resistive loss. The parallel connection of seven such rows through a screen-printed silver bus gave a solar cell module measuring 40 cm × 25 cm (0.1 m2). The active area was 65% of the total area. The remaining 35% of the area was used for interconnections between cells and for the separation between rows. The 65% active area was chosen to encompass a good margin for prototyping/research and to keep contact resistances between the cells low. In a fully automated process the active area could perhaps reach 90–99% interval but problems with current extraction and interconnections were found to become very critical. There are obvious shortcomings to this approach but the advantage of low current density is believed to be the biggest problem in efficient energy extraction from the module when no simple method for reducing the sheet resistance is available. In the simple geometry ITO/MEH-PPV/aluminium the module gave an open circuit voltage (Voc) of 10.5 V, a short circuit current (Isc) of 5 μA, a fill factor (FF) of 13% and an efficiency (η) of 0.00001% under AM1.5 illumination with an incident light intensity of 1000 W m−2. A geometry employing a sublimed layer of C60 (ITO/MEH-PPV/C60/Al) improved Voc, Isc, FF and η to 3.6 V, 178 μA, 19% and 0.0002%, respectively. The lifetimes (τ½) of the devices defined as the time it takes for the module efficiency to attain half of its maximum value were found to improve significantly when a sublimed layer of C60 was included between the polymer and the aluminium electrode. The modules were laminated with 200 μm polyethyleneterephthalate (PET) foil to mechanically protect the cells. τ½ values of 150 h were typically obtained. This short lifetime is linked to reaction between the reactive metal electrode (aluminium) and the constituents of the active layer. The modules were tested outdoors in different weather condition (wind, high temperature excursion, rain, snow). Tested during a storm the polymer photovoltaic laminate was subject to vibration stress and deformation and delamination in the organic layer was observed with fast bleaching of the active material. Efficient encapsulation with barriers that has very low oxygen and water permeabilities will be needed before future commercialisation can be anticipated.

Introduction

Polymer photovoltaics [1], [2], [3], [4], [5], [6] is a very promising technology that in principle offers the possibility for the production of solar cells at a fraction of the cost of expensive silicon based semiconductor photovoltaics in high volumes. The production of polymer based photovoltaics using industrial screen printing have demonstrated the possibility of producing on the order of 1000–100,000 m2 on a process line per day. The production of the same solar cell area based on silicon in a state-of-the-art production plant typically takes 1 year. There are however still some shortcomings of organic photovoltaics that need a solution before commercialisation at that level will be possible. The efficiency and especially the lifetimes of polymer photovoltaics are currently low. Efficiencies just below 5% have been reported [7], [8] on very small active areas and while this is getting closer to the range of efficiencies obtained for silicon based solar cells (10–25%) there is still some room for improvement. The module efficiencies for commercial silicon solar cells are in the 10–20% range and data for polymer solar cell module efficiencies are needed. Commonly efficiency data for organic photovoltaics are based on very small solar cells with active areas of few mm2. Research on large area organic/polymer photovoltaics is dawning and research on organic/polymer photovoltaic modules has not been addressed in any great detail. The purpose of this work was to address the problems associated with large area organic/polymer photovoltaics, printing and patterning of the active layer [9], [10], [11] and module preparation. The typical device structure for organic/polymer photovoltaic devices is a multilayer planar geometry with one or more active layers sandwiched between two electrodes. Commonly a substrate carrier bearing one of the electrodes is used for patterning the active layers that may involve a vacuum deposition process. Following on from this the final electrode is applied. Most often the final electrodes are applied in a vacuum coating process by evaporation of a metal like aluminium using thermal or e-beam evaporation. A process line capable of handling large area organic/polymer photovoltaics in a batch process has been designed for these research purposes.

In this paper we demonstrate the production of large area polymer based solar cells and encompass aspects of module design in the context of current polymer photovoltaic materials. We address the problems associated with the connection of many solar cells into a module and demonstrate a 0.1 m2 polymer solar cell module Fig. 1.

Section snippets

Active material

Poly-1,4-(2-methoxy-5-ethylhexyloxy)phenylenevinylene (MEH-PPV) was prepared by the Gilch route [12]. Large amounts were needed for the screen-printing and it was found impractical to use volumes larger than 10 L. Ten batches were prepared each starting from 200 g of monomer. This gave 600 g of crude MEH-PPV. All the batches were combined and dissolved in THF (20 L) by careful mechanical stirring under argon and with the exclusion of light for 48 h. The thick solution was then filtered through glass

Results and discussion

The active layer of polymer based solar cells has mostly been prepared by spincoating, casting or doctor blading and while these techniques are excellent for fundamental studies large area high volume production is anticipated to require different techniques like roll-to-roll coating, spraying or screen-printing. Screen-printing has been employed for the printing of the active layer of polymer photovoltaics in small area [9], [11] and large area [10]. These promising results led us to attempt

Conclusions

The design and fabrication of 0.1 m2 solar cell modules was demonstrated and pertinent issues in relation to interconnection, lifetimes and mechanical stability was detailed. It was found possible to achieve an active area of 65% of the total module area. The efficiencies obtained for the modules were much lower than expected. Possible reasons for this behaviour is believed to be contact problems, atmospheric handling during device preparation and possibly electron radiation damage to the active

Acknowledgements

We would like to thank Niller Fagerberg for technical assistance with the large scale synthesis of MEH-PPV. This work was supported by the Danish Technical Research Council (STVF 26-02-0174, STVF 2058-03-0016), the Danish Strategic Research Council (DSF 2104-04-0030) and Public Service Obligation (PSO 103032 FU 3301).

References (14)

  • H. Spanggaard et al.

    Sol. Energy Mater. Sol. Cells

    (2004)
  • C. Waldauf et al.

    Thin Solid Films

    (2004)
  • C.J. Brabec

    Sol. Energy Mater. Sol. Cells

    (2004)
  • F.C. Krebs et al.

    Sol. Energy Mater. Sol. Cells

    (2004)
  • B. Winther-Jensen et al.

    Sol. Energy Mater. Sol. Cells

    (2006)
  • F.C. Krebs et al.

    Sol. Energy Mater. Sol. Cells

    (2005)
  • C.J. Brabec et al.

    Adv. Func. Mater.

    (2001)
There are more references available in the full text version of this article.

Cited by (257)

  • Geometrical optimization for high efficiency carbon perovskite modules

    2019, Solar Energy
    Citation Excerpt :

    The poor conductivity of the FTO and carbon electrodes limits a further performance improvement for large scale C-PSC modules and requires careful design to overcome. A ‘sub-cell design’ is used with series connection of individual cells to minimise resistance related losses (Krebs et al., 2007). This ‘sub-cell’ design has to maximise the dimensions of the active area and minimizing the dimension of the non-active area, including the interconnection area, whilst still minimizing the resistive losses.

View all citing articles on Scopus
View full text