Abao KUBU residency

The development of DSSC starvation synth

• 
• 
• 
• 
• 
• 

Few previous researches has been made before this residency in Taipei with Marc Dusseiller in February. This 5-weeks residency at KUBU, is part of the The Garden And The Hedge Summer Exhibition And Program, organized and curated by Teresa Dillon and Tuomo Tammenpää. This research aim to develop graphical series connection in DSSC to satisfy both artistic and electric engineering purposes; the challenge is to have graphical TiO₂ and platinum electrodes without failing it's electric properties and efficiency. This roof of concept prototype is expected to generate more complicated pattern base on the voronoi fractal foundation in the future. The graphics could be also engineering useful too, for example, this POC could be useful for further researches, such as self-powered robotic eyes, smart contact lens or sunglasses in the future due to the fractal series circuits could corporate with the micro lens array in the focus-free lens system to exploit DSSC in the manner of smart visions.

• 

  The opening of Garden And the Hedge 2025 at KUBU.

• 

  The opening of Garden And the Hedge 2025 at KUBU.

• 

  The opening of Garden And the Hedge 2025 at KUBU.

The design of graphical W series circuits in DSSC

There are several different configurations of DSSC, the configurations defines the architecture of series or parallel circuit in DSSC, or even series and parallel. The work of etching conductive glass is crucial for the design of the configuration. During the KUBU residency I choose W-type over Z-type becuase it doesnt requires the participation of silver traces. In order to power a synth board which requires at least ~1.2V and 0.12mA. In general, a single DSSC approximately harvest 0.6V/15mA with active area in 3x3cm, therefore we need to series at least 2 sub-cells within the one DSSC. Eventually two array designs were made, 6x7 and 4x4.

Additionally, another interesting feature of the W-type architecture is that it distributes the TiO₂ electrodes (the semiconductor material that converts photons into electrons) across both FTO glass substrates. These two FTO glasses can be immersed in different dyes to produce different colors. As a result, their light absorption abilities differ, which makes the generated sound also 'two-sided'.

• 

  Reference of diagram and the basic principle of Z and W series circuits in DSSC.

• 

  The graphicalized etching pattern and electrode sequences of the two electrodes of a W-type DSSC.

Graphical electrodes made with voronoi generative patterns

Two working poc prototypes were made in voronoi array during the residency, proving that all sub-cells in DSSC are successfully stakced together to create high voltage. one in array of 6x5 and the other one in 4x4. The voroni patterns were generated in Max/msp with jit.gl.bfg object, there are several reasons to use voronoi in DSSC configuration design. Firstly, due to the current output will be lower averagely by the smallest sub-cells in both parallel and series circuit, voroni contribute an interesting fractal structure while keeping all electrodes in similar size at jitter param of the jit.gl.bfg is below 0.14. The other convenient params of the voronoi function is voronoi_crackle, it is a quick way to define the distant between each sub-cells, the etching pattern, the sealant pattern. The distant between each-sub cells is also relevant to the shunting level within the DSSC when working with a non-sealant cell.

• 

  A simple max/msp patch to generate voronoi pattern to meet the requirement of the design of the screen printing for DSSC electrodes.

• 

  The voronoi_crackle pram of the jit.gl.bfg adjust the distant between every sub-cells.

• 

  The design of the electron pathway in a graphical W series circuits in DSSC, the index numbers indicates the sequential order in which the electron travel through the TiO₂ (red) and platinum (black) electrodes in two FTO glasses. The arrows indicate the beginning cell and the end cell in the circuits.

Etching FTO with CO2 laser

In order to create w-type we need to etch the FTO layer properly in order to series all sub-cells together. The etching was made with a xTool P2 laser cutting machine in KUBU. The working spectrum of CO2 laser is at 10600nm which is effective to FTO, the conductive transparent layer of the conductive glass. The params of the laser is set to engrave mode, 15% of power, 50mm/s, pass 1, 50 lines per cm.

• 

  Etching pattern of the DSSC with 6x7 sub-cells in voronoi pattern.

• 

  The engrave mode is set to 15% of power, 50mm/s, pass 1, 50 lines per cm. The cut mode for thermoplastic sealant is set to 10% of power, 100mm/s, pass1.

• 

  Close up photo of the CO2 laser etched FTO glass in 60x66mm.

Preparing screen printing masks

The screen printing masks can be DIY in KUBU. The frame was prepared with photosensitive emulsion coated with a paper blade, and dried in dark room over night. The photographic transparency is printed with EPSON ET-7750 inkjet. The exposure light source is two 20W UVA tubes, exposure duration is ~7.5 minutes. Once the exposure is done, rinse a way the uncured part in the water tank.

• 

  The screen printing UV exposure process.

• 

  The screen printing station in KUBU.

• 

  UV lamp spec.

• 

  Screen printing emulsion used for mask.

• 

  The screen printing frame and the emulsion blade in KUBU.

• 

  Closeup of screen printing mesh of TiO₂ electrode.

The coating of TiO₂ and platinum paste

Typically a complete DSSC takes 2 to 3 prints to complete, TiO₂, TiO₂ blocking layer (optional, for increasing the current) and platinum paste onto one FTO glass. The two glass electrode was coated with 18NR-T TiO₂ and platinum paste purchased from Greatcell Solar. It is quite difficult to land the pastes precisely mapped with the etching pattern on FTO glass without a transparent silk screen mask, a simple printing station made of a small led panel and a sheet of normal glass to make the mapping work much easier. After the printing, there are still enough space between each sub-cells for the surlyn film to be put in.

• 

  A simple DIY screen printing station with backlight and glass platform.

• 

  The two electrodes placed on top of the screen printing station with backlight.

The sintering

After coating the two electrodes, the two electrodes were sintered with a household stove which is easily find in super market, there are usually 6 degrees of it, set it to 5 usually produce 380ºC to 420ºC in open air. The sintering is to evaporate the solvent and making the TiO₂ nano particles to form the nano porous structure for increase the space for more dye molecules to be attached in. The duration is completed around 30 minutes, or you can observe the color change of the TiO₂ to define one complete bake cycle; when the TiO₂ is heated up over 260ºC, its color starts to change from white to dark brown and goes back to white again after reaching 400ºC, indicating a sufficient sintering cycle is done. Staying at 420ºC for about 10 minutes more and then turn off the stove to let it cool to room temperature.

• 

  The TiO₂ and pt electrodes being sintered with electric stove.

• 

  The sintered and etched electrodes.

The dyeing and the two sided photosensitivity

The efficiency of a DSSC is defined by the dyeing process, for example, the duration, the density of the dye, the type of the dye. In this case, I firstly dye the electrode with N719 dye purchased from Greatcell Solar, and use natural dye to tune its' color after. The dyeing duration takes at least 2-4 hours to have sufficient current to power a starvation synth. If only natural dye is used, it requires at least 8 hours or overnight dyeing to power a starvation synth board.

Since the W-type architecture distribute TiO₂ region on both glass, so we can soak the two glasses into different dye to have a bi-color cell. Artistically it enrich the outlook, electrically it defines the two electrodes with different ability to absorb light, therefore it is also useful for synth design, because two sides of glasses sound "two-sided" when facing to the sun.

• 

  The two electrodes of w-type DSSC were soaked into N719 dye and hollyhock dye.

• 

  N719 and Hollyhock dyed DSSC electrodes.

Sealing with Surlyn film

After the dyeing process is done, take the electrodes out of the petri dishes and rinse away the extra dyes with ethanol. I don't have techniques nor tools to drew holes on glass without breaking the glass, also it doesn't make sense to drew holes for every sub-cells in w-type cell. So I need a alternative way to inject the electrolyte after the electrodes are thermal sealed without drilling any holes in glass electrodes. The thermoplastic sealants used in this process has melting point of 90ºC, thickness in 10um. I decided to separate the laser-cut sealants into several parts to leave open gaps. The several laser-cut sealants are carefully placed onto the electrode one by one by tweezers. Once its done, check if all sealants are in alined with the etching patterns of the FTO glass. Finally, put another electrode on top of it with the coating side face down, to sandwich the sealant without moving or offset it, fix the sandwiched electrodes with clips. The clipped sealants are then heat up with heat gun to ~140ºC for 30 seconds and are cooled to room temperature naturally.

Bonding two electrodes significantly increase the life span of the cell from keeping away the drying electrolyte issue. However the 60x60mm cell is quite big and the deformation caused by 450ºC is not avoided, the FTO is often banded after sintering and therefore causing the difficulty to seal the two electrode good manually.

• 

  Surlyn film being used for sealing DSSC with in 10um thickness and melting temperature at 90ºC.

• 

  Peeling the laser-cut surlyn film, the shape is generated with voronoi.

• 

  Laser cut surlyn film is sandwiched in between two electrodes.

• 

  Clipped and sandwiched two electrode.

• 

  Bonding up the clipped electrodes with heat gun.

Electrolyte injecting

After the two electrodes were bonded together with Surlyn film, the iodine liquid electrolyte was injected at the edge of the glass electrodes and drawn in through the sealant opening by capillary action. The electrolyte travels good within the 60x60mm area without problems, indicating the success and good quality of the sealing work. Usually I use the EL-HSE electrolyte purchased from Greatcell Solar. The content of the donated electrolyte here is I₂: 0.05 M, NMBI (benzimidazole): 0.3 M, GuSCN (guanidinium thiocyanate): 0.05 M, PMII (1-propyl-3-methylimidazolium iodide): 0.8 M. Solvent: Mixed solvent of MPN (3-methoxypropionitrile) and PMII in a volume ratio of 25.5:74.5 (v/v).

• 

  DSSC electrolyte donated by Janne Halme.

• 

  The iodine electrolyte injection for DSSC array.

3D printed enclosure and the lid

There are two parts of enclosure, the lid and the case. The assembled cell is secured inside a 3D-printed enclosure (printed on a Bambu Lab P1S) with minimal design. The 4 edges of the assembled cell are applied with UV glue, and cured with the same UV lamp in KUBU screen printing room for one minute. This could be helpful to seal the openings left in the sealant layers. The two piezo buzzers for non-linear sound generation in starvation circuit are exposed at the back of the enclosure to highlight the key feature of starvation synth circuit. The black enclosure is made of Bambu PLA Basic filament, the white enclosure is made of ESUN PETG filament with a bit higher printing temperature 230ºC - 250ºC. The diameter of filament are both in 1.75mm.

• 

  Bambu lab P1S and the printed englosures.

• 

  UV glue used to bind cell into enclosure, purchased in Tokyo Hand.

• 

  The front view of the finish enclosure with DSSC and synth board installed.

• 

  The back view of the finish enclosure with DSSC and synth board installed.

Voc and Jsc measurement

The Voc and Jsc of the two w-type DSSC array was measured with simple multimeter. The Voc and Jsc of the 6x7 array with voronoi shape has reached 15V and 0.41mA. The 4x4 array one was measured ~4V/0.8mA right after the cell was manufactured. These values are still under the expected theoretical values but already good enough as proof concept prototype. However the efficiency drops after few days, it is maybe caused by dye bleaching or the improper sealing cased shunting.

• 

  Voc of w-type 6x5 array DSSC.

• 

  Jsc of w-type 6x5 array DSSC.

Design of starvation synth board

The design of the board in this case is designed by Marc Dusseiller. The relevant files can be found here. The design is inspired by the project of the starvation synth made by Ralf Schreiber. Starvation circuit is a low power consuming circuit that generate bizarre noise with low current, that's why it is a good option to embody the weak power harvest by the DIY DSSC.

Basically there are 3 resistors and 3 capacitors in the circuits, but because of the concern of the minimal design in the project, symbolistically we only leave one potential meter knob on the board as "threshold" to starve the chip, although it is a bit odd since the DSSC it self is a photo-resistor that controls the light input. We should have connected the DSSC array to the other resistors which are relevant to the sound mixings. Also the "two-sided" photosensitivity was not included in the synth design, but much more to be upgraded in the future.

• 
• 

Connecting starvation synth circuit with RAVE

This is an alternative application of starvation synth, and a little more introduction for RAVE developed by Ircam. I documented it here Connecting starvation synth circuit with RAVE.

Workshop at Synthcamp 2025

A small DSSC workshop at Synthcamp!

• 
• 
• 
• 
• 
• 

A small DSSC gathering with Janne Halme in Aalto University

I found chemical leaking and lost all my iodine in my suitcase after my flight landed in Helsinki. So Marc connected me to Janne Halme, a lecturer in Applied Physics in Aalto University, to seek for some helps. Surprisingly he donated so many chemicals to me and gave me a small tour of their DSSC laboratory and their previous art project Towardless: Plant-based Electrical and Art Energy in 2021. I am very thankful for all of the donations, I couldn't accomplish this program without these valuable donated materials during my staying in KUBU.

• 

  DSSC art of Janne Halme.

• 

  Sun-powered Textiles by Janne Halme.

• 

  Donated DSSC chemicals from Janne Halme.

• 

  A small DSSC meetup in Aalto and the signed paper to transfer the responsibility of chemical disposal to me.

• 

  A silicon solar cell art gift from another research group to Halme Janne.

• 

  A close-up photo of the DSSC artwork "Blck Vlvt", co-created by Bart Vandeput, Janne Halme, and Pyry Mäkinen. Photo by Anne Kinnunen.

References

1. Giordano, Fabrizio, Eleonora Petrolati, Thomas M. Brown, Andrea Reale, and Aldo Di Carlo. 2011. “Series-Connection Designs for Dye Solar Cell Modules.” IEEE Transactions on Electron Devices 58 (8): 2759–64. https://doi.org/10.1109/TED.2011.2158216.
2. Seo, Hyunwoong, Minkyu Son, Jitae Hong, Dong-Yoon Lee, Tae-Pung An, Hyunju Kim, and Hee-Je Kim. 2009. “The Fabrication of Efficiency-Improved W-Series Interconnect Type of Module by Balancing the Performance of Single Cells.” Solar Energy 83 (12): 2217–22. https://doi.org/10.1016/j.solener.2009.09.003.
3. Miettunen, Kati, Piers R.F. Barnes, Xiaoe Li, ChunHung Law, and Brian C. O’Regan. 2012. “The Effect of Electrolyte Filling Method on the Performance of Dye-Sensitized Solar Cells.” Journal of Electroanalytical Chemistry 677–680 (July):41–49. https://doi.org/10.1016/j.jelechem.2012.04.013.
4. Shih Wei Chieh. 2025. 1, 540, 000nm Of DSSC. http://archive.org/details/1-540-000nm-of-dssc-in-tokyo.
5. https://github.com/shihweichieh2023/w-type-DSSC