Skip to main content

Testing and up scaling of technology developed under the SOLNOWAT FP7 Project which developed a competitive 0 GWP, dry, atmospheric pressure etching process for use in manufacture of PV solar cells

Final Report Summary - DEMOSOLNOWAT (Testing and up scaling of technology developed under the SOLNOWAT FP7 Project which developed a competitive 0 GWP, dry, atmospheric pressure etching process for use in manufacture of PV solar cells)

Executive Summary:
DEMOSOLNOWAT aims to demonstrate at an industrial scale the viability of a dry process alternative for the Solar Photovoltaic cell industry.

Following on the successful results of SOLNOWAT FP7 project that commenced in September 2011, this project focus is on up scaling and demonstrating the novel technology that was developed for the photovoltaic manufacturing industry. The technology developed and proven in the SOLNOWAT project serves to reduce the water consumption of the cell manufacturing process, focussing on the etching steps, high throughput technology and specific process control - this follow on project act as a stepping stone in getting the innovations from the "lab to the fab".

Under this project specific industrial scale prototype equipment were built by the participating SMEs and the technology demonstrated and tested The prototype equipment features a dry etching process line with high volume/ high throughput capacitry, integrating specific process control.

Equipment Prototypes were built and scaled, and tested in industrial conditions:

- Automated Atmospheric Dry etching/texturing tool prototype with capacity of 1000 wafers/hour
- Integrated non-contact wafer conveyor
- Industrial F2 gas monitoring mass spectrometer hardware prototype

This project allows the SMEs to demonstrate the viability of their new technology and process equipment to the Photovoltaic industry at an industrial scale, and bridge the gap to market. The consortium is composed of 3 SMEs located in 3 different European countries.

Project Context and Objectives:
DEMOSOLNOWAT aims to demonstrate at an industrial scale the viability of a dry process alternative for the Solar Photovoltaic cell industry.

The current water usage in the photovoltaic (PV) solar cell manufacturing industry is not sustainable.

The PV solar industry as a whole has been growing dramatically over the last 10 years. PV solar is indeed recognised as the renewable energy alternative that can meet our global energy needs for the foreseeable future. Companies involved in this industry have been growing significantly, answering the demand, and scaling their production capacity accordingly. Equipment and processes developed by cell manufacturers have been for most adopted and scaled up from semiconductor manufacturing.

However, the value-add and cost structures for both sectors are vastly different despite the manufacturing processing technologies being similar. As the industry continues to increase its capacity, very large footprint factories that have heavy consumption of chemicals, water and emissions of high Global Warming Potential (GWP) gases become unsustainable. New regulations, including the Kyoto agreement, will enforce strict control on water management and emissions. The availability of environmentally friendly production technologies that can cope with emission regulations in Europe will be crucial for the continuity of cell manufacturing in the EU. At the same time EU regulations are expected to further restrict the use of production technologies with high GWP.

There is therefore a need to rethink and develop new process solutions that meet these requirements in order to strengthen the economic and ecological innovation ability of European SMEs and cell producers.

This demonstration project applied the results and processes successfully developed and demonstrated during the SOLNOWAT project. The wafer texturing technology developed allows for the reduction in the very high water consumption and Global Warming Potential (GWP) emissions of the current manufacturing process while meeting all industrial production requirements, and improving the efficiency of the resulting solar cell device.

Equipment Prototypes were built and scaled and tested in industrial conditions:

- Automated Atmospheric Dry etching tool prototype with capacity of 1000 wafers/hour
- Integrated non-contact wafer conveyor for wafer handling
- Industrial F2 gas monitoring mass spectrometer hardware prototype for process control

This project will allow the SMEs to demonstrate the viability of their new technology at an industrial scale to the Photovoltaic industry, and help them bridge the gap to market.

Explanation of Existing Processes for Solar Cell Manufacturing:

Silicon etching is a key technology in various processing steps during the production of PV solar cells:
- Removal of sawing damage that has occurred during the “wafering” process,
- Texturing of the Si wafer to reduce its reflection prior to the formation of the emitter layer.
- The removal of the residual phosphorus silicate glass (PSG) that grows during the high temperature emitter formation step.
Currently most of the etching steps are carried out using wet chemistry equipment; the wafers are moved across large baths containing chemical liquid mixtures. Large volumes of water are also required for rinsing after etch steps, typically 5 litres/Watt. It takes about 1.5L of “tap water” to produce one litre of DI water.

1 GW factory requires 5,000,000,000 Litres /year = 9,645 litres/minute of DI water= 14,467 litres/minute of “tap” water

A 1 Giga Watt (GW) factory uses up to 15 000 litres/min of water. Theses staggering numbers outline how unsustainable wet processes are, if the PV solar industry is to grow as predicted. Batch style systems require numerous wafer handling steps and special dedicated carriers. Wet
technology is not versatile, and therefore the equipment has to be designed according to the type of wafer to be processed, reflected by the range of offers.

Technical limitations of existing wet process:

- Unsustainable due to water consumption
- Poor process control
- Limited throughput (due to large footprint)
- Very large tool footprint on factory floor (up to 17m long)
- Not adapted to advanced cell concepts or thin wafers
- Not truly single sided process
- Process recipe highly dependent on substrate used
- Relatively poor texturing efficiency for mono-crystalline wafers for inline wet chemistry or longtime process for batch alkaline (anisotropic) texturing
- Surface contamination issues after a wet clean and subsequent drying

DEMOSOLNOWAT focuses on dry atmospheric pressure silicon etching technologies using non GWP gases to achieve the precise textures required for efficient solar cell manufacturing. The technology incorporates a high performance process monitoring solution and high added value non-contact wafer handling technology.
To be adopted by the cell manufacturing industry, the processing equipment needs to be robust and will have to demonstrate strong controllability and stability in industrial conditions.

The technology developed provides a process that is compatible with future solar cell technologies (advance cell concepts, the processing of thin wafers and incoming environmental regulations) while also providing cost reduction. These are essential requirements for a successful commercialisation.

This project follows the development strategies outlined by the international SEMATECH Manufacturing Initiative5 in order to reach a targeted 50% reduction in Green House Gas emission for a future PV system.

- Improve process optimization
- increase the overall gas utilization
- use non GWP gases chemistry, like F2 (GWP of 0)

Present day technology allows cost-effective abatement with limited emissions, but the roadmap to large scale PV production will target zero-emission policy. This is in line with the technology used in this project. The European Commission took an important step this year towards long-term climate objectives by presenting a proposal to significantly reduce emissions of high GWP fluorinated gases
in the European.

Project Results:
ALYXAN - Mass Spectrometer prototype:

The SOLYZE mass spectrometer instrument has been designed and built in order to answer the specifications of the DEMOSOLNOWAT project. The instrument includes a VQM analyser which has been fitted to a vacuum chamber based on a 6 way cross, a pressure gauge, two primary pumps (one of which is chemical), a sampling box, an electronic board and a computer. Two parts of the instrument (the 6 way cross and the sampling box) can be heated and regulated in temperature to avoid absorption on surfaces and for a good stability of the analysis. It is possible to sample gases 5 meters away of the instrument and the use of corrosive gases is not a problem thanks to the use of a chemical primary pump and a sealing gas in the turbomolecular pump. The instrument can be used directly with a keyboard, a mouse and a screen connected on the integrated PC or it can be controlled from another computer by a remote connection.

Sampling part of the instrument has been optimized in order to avoid dead volumes and therefore decrease the response time of the instrument. The sampling part is the most demanding part of the instrument, it has to be set up properly to achieve the best performances. A double drop pressure configuration is used in order to correctly sample the gas. This double drop of pressure is obtained by the use of a capillary and a metering valve enabling a precise control of the pressure inside the instrument. The length of the capillary has been optimized thanks to tests and a length of 5 meters -with an internal diameter of 0.5 mm- was found to be the best option for the application. A new valve is considered in order to simplify the sampling part of the instrument. This new valve will be able to sample a gas from atmospheric pressure to UHV without the need of other valves or a primary pump.

A software (running on a PC under the Windows operating system) has been developed for the instrument allowing to control the different elements of the instrument (valves, pumps, heating...) and to gather and display the data obtained by the analyser. Some automatic procedures have been created in the software in order to simplify the use of the instrument. Those procedures allow to start/shut down the instrument, and optimize the different parameters of the analyser. Following and identifying the peaks obtained are possible by using a library that has been created in the software and can be completed easily. A calibration curve can be added for each molecule, allowing to quantify the species if needed. The data are displayed thanks to a trend graph as well as a mass spectrum and can be saved on excel sheets. Securities have been set up on the heating and the valves in order not to degrade the instrument.


The loader/conveyor/pre-heat system uses the non-contact levitation technology with ultrasound. For the use of non-contact handling, the sonotrodes consist of an aluminium-alloy plate using the physical effect of suspension for flat substrates. These plates are designed in a specific shape for an equal vibration pattern causing equal levitation forces to the substrate or solar wafer to keep in a distance of approximately 100 to 180 µm.
The wafers are provided by a wafer loading system with a buffer system, preheated zone with three heating zones, a process chamber/etching zone with an integrated purge curtain on the feed in and on the output. Followed by a cool down track with two temperature zones and an unloading system(Multigrippertechnology). Loading and unloading system are designed for a continuous running wafer flow.

All feeding (unloading) modules 1 and 2 are designed in a modular design. Each module can work indenpendently. The communication between the modules is based on “Ethercat”. For the wafer handling control is a PLC from Beckhoff used. The wafer handling system can be controlled by a HMI or by the process module or a “MES-System”.
The modules 2 and 6 deliver a homogeneous heating respectively cooling of the wafers and have non- contact handling to avoid any hot or cold spot during the transport. They are also designed for a modular set up and a simple line integration.All modules have a emergency stop circuit which interrupts the dangerous movements and the heating. The emergency stop of the handling system can be coupled with the Nines ADE module. The Loader with buffer containing 50 wafers supply continuously an endless flow of solar wafers to the Nines ADE process module. This allows the machine to run semi-automatically with the only intervention needed from the user to change the cassettes. The addition of the different buffer units allows the Nines ADE system to etch on a continuous basis. The Loader assembly has a capacity for 2 cassettes of wafers, which are loaded at the higher level into the loader. On the lower level there is also a cassette buffer with a capacity of 2 cassettes.

The operator is loading the J&R Cassettes full of wafers onto the loader cassette buffer equipped with a conveyor. The conveyor moves the cassette to a cassette lift. The Cassette lift is moving down slot by slot to allow the wafer feeder to remove a wafer. When all the wafers are removed from the cassette, the cassette drops to the lower level and is transferred out by the outgoing cassette buffer conveyor for the user to remove. The wafers are transferred from the cassette to the wafer buffer by a wafer feeder, this wafer feeder is also used to transfer wafers from the buffer into the pre-heat module.
Buffer module: it allows time to exchange cassettes without affecting the continous wafer supply for etching. The wafer buffer has a capacity of 50 wafers. By feeding one wafer per second, the operator has a time frame of about 45 seconds to change the cassette.

In the Pre-Heat Zone the Solar PV wafers are moved to 3 Sonotrode zones which are heated by Ceramic Heaters. It is essential to pre-heat the wafers to controlled temperature levels so that the wafers are not exposed to thermal shock. The temperature of each sonotrode is controlled by a closed-loop control system with heater control and infrared temperature sensors.In the Cool down zone the Solar PV wafers are moved to 2 Sonotrode zones which are cooling down the wafers controlled by Ceramic Heaters. It is essential to cool down the wafers to controlled temperature levels so that the wafers are not exposed to thermal shock.

The unloading system consists of two X-Z axis with one “Multi gripper” respectively. One Multi gripper is able to take 25 wafers using the edge grip principle, i.e. the wafers are touched only at the outer edges, avoiding surface contact and contamination as far as possible. While one Multi gripper is collecting the wafers from the transportation system, the other one is carrying the wafers to the cassette. Due to the Multi gripper it is possible to receive wafers from the conveyor during the cassette exchange. So this allows a continuous wafer flow to or from the process without any interrupt for cassette exchanges.

The operator is loading J&R Cassettes on the system front side to the lower cassette conveyor, the conveyor moves them to the wafer loading station. On the wafer loading station the cassette gets clamped in a fixed position, so that the wafer slots can be filled by the Multi gripper. After four filling processes the cassette is completely filled with 100 wafers and can be moved out by the lower conveyor. The user can remove the full cassette on the backside of the system.The Multi gripper has a capacity of 25 wafer per catch. There are two side locks to avoid relative movement of the wafer against the fork surface during the movement.

Overall, velocities of > 1000 wafers/hours were achieved for this single lane system.


For the purpose of this demonstration project, it was proposed to build a single lane prototype offering a throughput of >1000 wafer/h. This prototype integrates wafer handling solutions and process control hardware. Because the technology is modular and, higher throughput tools of up to 4000 wafers/h required for 120 MW factories can be built using several such lanes in parallel, offering a straightforward way of scaling the technology further, although this is beyond the scope of this project.

The reactor that compose the core of the system was design including a set of purge curtains.The function of the Entry side Purge and Purge extraction module is to ensure no contamination enters the Etch zone within the reactor and that no Etch gases or by products exit the reactor. The purge curtain is achieved by flowing nitrogen through small orifices creating a vertical plane of nitrogen that wafers pass through. The purge extraction system removes the nitrogen and any contamination which may have entered on wafers.
To process wafers, the Reactor module runs at temperature between 150 and 350° C and within the reactor module the etch gas reacts with solar pv silicon wafers removing approximately 2um of material from the top surface. The etch process changes the wafer surface characteristics, which means the wafers absorb more sunlight, which produces higher efficient solar cells.
The reactor design provides an enclosed sealed solution where the F2 gas, nitrogen and process by products are extracted from the system by the facilities extraction system and by using a series of nitrogen air curtains and extraction channels, the air curtains eliminate the process gases from travelling out of the reaction zone or travelling out with the product. The flow of purge gases within the reactor and the extraction pressures generated are critical to the etch process efficiency. The wafers are transferred through the reactor on a conveyor belt, where the wafers are held by vacuum on the heated belt. The etch conveyor is contained within a controlled nitrogen environment and a thermal environment of above 150 degrees. The reactor body was redesigned to allow access to the Gas Delivery Plates area without the need to open the reactor assembly from the underside. The GDP is attached to the conveyor assembly in a stacked design manor. This allows the GDP to be changed quickly to create different process conditions. The etch gas is supplied from the Reactive panel which is contained within an enclosure on the rear side of the machine. The gas for the series of air curtains is from the inert panel.

The inert panel forms the first part of the machine gas delivery system. The panel is mounted within the Nines machine ventilated enclosure. The main function of the inert panel is to provide Nitrogen to the Etch system air curtains. The air curtains are the most critical part of reactor operation, they prevent in conjunction with the extraction system, the F2 gas and the process gases from escaping to the machine enclosure environment.

The reactive panel forms the second part of the machine gas delivery system. The Fluorine gas is delivered to the panel in a stainless steel coaxial pipe. The Fluorine lines are shut off using double isolation metal sealed valves. The nitrogen supply to the panel is delivered in a ½” Stainless steel line reducing down to ¼” at the connection point. Each of the Nitrogen lines has a digital pressure switch and independent line filters. The reactive panel has four MFC’s, two for the fluorine lines and two for the nitrogen lines. The panel is mounted in a ventilated enclosure within the machine enclosure which is extracted by the facility extraction system.

The PV wafers are held on the conveyor belt by using a vacuum. The vacuum is generated by venturi and travels through a series of small holes in plates under the conveyor belt. The conveyor belt is kept flat by using a vacuum. The wafers are held in position on the conveyor belt through a series of small features cut through the conveyor belt which allows the vacuum to the underside of the wafer. Underneath the vacuum plates there is a series of element heaters which heat the conveyor belt and the wafers. The heaters are connected in groups which form a temperature controlled zone.

The laboratory setup at NINES allows to process safely wafers using Fluorine. A special gas cabinet and associated automated controlled was commisioned and located next to the lab. A life safety system was put in place, including sensors, alarms. All tools were designed to "fail safe" standard, i.e. all valves would closed when powered down. A point of use process scrubber is used next to the tool in order to treat the exhuast gas immediatly and remove any harmfull susbstance. A powerfull ventilation system was put in place in order to evacuate the equipment enclosures as well as providing lab and F2 room with refresh rate that provide good safety levels.

Potential Impact:
The SMEs partners of DEMOSOLNOWAT have focused on innovation in the manufacturing process and will gain a competitive edge by offering added value through a processing solution that is easily integrated with current manufacturing process flows.
In the PV solar cell manufacturing industry the demonstration of a novel manufacturing process is vital for the acceptance by the manufacturing audience as a whole. As all steps in the manufacturing of a PV solar cell are interdependent, any change to one single manufacturing process, no matter how small, can take a long time before being qualified or passed for volume production in a high output cell manufacturing plant. In this “Design-in” Phase (a change is being designed into a full manufacturing process) the new process is run as it would be in volume production.

Large amounts of statistical data are collected for a number of different process parameters - following careful analysis of the data the cell manufacturer can say with confidence that the process is stable and ready for volume production in its plant. Starting this “Design-in” process is a very large commitment for a cell manufacturer and they will not start it without a level of confidence that:
1. The process will work for their requirements
2. The process solution provider can deliver the required solution for full volume production (may be 10-15 units) on time and in budget
3. The solution provider can provide on-going support right through the design-in phase and on to the volume production phase for their operations.

The SMEs will use the demonstration activities in the DEMOSOLNOWAT project to create the level of confidence that a cell manufacturing customer needs. The dissemination of results from this project is a route to market for the “total solution” offering for volume production.

Throughout the whole project many activities involved attending conferences and meeting with potential customers – spreading the word about the “total solution” to be demonstrated under the DEMOSOLNOWAT project. In the second phase of the project the consortium led by Edward Duffy of Nines PV started to engage with Tier 1industrial customers mainly outside of the EU. The here was to try to counter the trend of PV manufacturing shifting eastwards and to make sure that the EU SMEs offer their product to this very important Tier 1 customer base.

In terms of disseminating DEMOSOLNOWAT to industry, the aim is to promote the uptake of the technology by all potential industrial stakeholders. As a result, the core message is hinged on the Unique Selling Points (USPs) of DEMOSOLNOWAT and the advantages/benefits that they will bring to the Solar PV manufacturers. Under the SOLNOWAT project we advised industry of the technology and the concept that we had proven – under DEMOSOLNOWAT we have engaged directly with Tier 1 industrial customers and have started Demonstration and evaluation process steps with many. The key elements that this cohort of customers has been keen to evaluate are:

• The ability to meet the needs of industrial production
• Actual Cost of Ownership data as opposed to extrapolations based on research work
• Yield data
• Reliability and suitability of the technology for industry i.e. services needed and new infrastructure.

For the Research Community, the key message is that there is an alternative viable etching technology that facilitates the use of advanced cell processing techniques at scale - as required for solar cell manufacturing in the 21st Century. The consortium needs to convince the research institutes/universities to adopt and work with this new technology as many of these have strong ties to and influence over industrial players. Communications with this audience needs to make clear comparisons with the incumbent and existing technology.

For the public, the key message focuses on the socio-economic impact of DEMOSOLNOWAT particularly the potential impact in terms of raising the profile and viability of PV solar for electricity production. The communications will be in plain language and without unnecessary technical jargon that can be confusing or counterproductive. The aim is to stimulate interest in DEMOSOLNOWAT and provide evidence of the need and possibilities for new and advanced technologies to be used for PV solar cell manufacturing, as well as to transmit the social benefits that can be derived from public funding.

In short, each target group requires their own tailored message, which is evidenced by the DEMOSOLONOWAT Communication Strategy.

List of Websites: