CORDIS - Risultati della ricerca dell’UE
CORDIS

Fibre Reinforced Composite Reflectors for Concentrated Solar Power Plants

Final Report Summary - COMPOSOL (Fibre Reinforced Composite Reflectors for Concentrated Solar Power Plants)

Executive Summary:
The COMPOSOL project developed a roadmap for the production of composite parts and innovative reflection surfaces for Concentrated Solar Panels (CSP).
The main idea was to replace the metallic structure of the CSP with a composite equivalent which will be lighter and hence require less energy (operational torque) to move, has improved stiffness in order to maintain the parabolic geometry of the reflector surface and be easier to maintain.
The design of the composite followed a modular architecture. This enabled the breakup of the whole structure into smaller components which are easier to manufacture, repair and replace. The accompanied structures in order to assemble the individual components became part of the design. Dedicated configuration management software has been developed in order to define each part, its position in the overall structure and the corresponding assembly.
The tools for the manufacturing of the composite part have also been designed.
The reflective surface was developed using Physical Vapour Deposition (PVD) of Ag onto an intermediate substrate (made from polymer or glass). The intermediate layer was then adhesively bonded to the composite structure. Using this approach, the technical difficulties of depositing the coating directly to the composite have been alleviated.

Project Context and Objectives:
The project had two main objectives:
• To establish a new concept of parabolic trough collector with a higher overall structural stiffness for a lower weight and
• To develop a high reflective surface, either directly on the composite structure or on a polymeric film that would be later bonded on the parabolic surface.

These developments will drive several secondary benefits such as:
• Lower maintenance cost due to a better environmental resistance
• Lower induced operational torques because of the lower weight of the structure
• A reduced power drive system as a result of the lower torques necessary for the movement of the structure.
• Faster movement to safe positioning in case of high wind speed.
• A more compact design, avoiding backing lattice structures

The tasks leading to the achievement of the two main objectives described above are the following:
A) Design of fibre reinforced components for the parabolic trough
B) Manufacturing process for the production of the reinforced components
C) A novel highly reflective and protective coating for CSP
D) Coating process for application to the composite structure
E) Bonding layer between prepreg and aluminium layer
F) Validated concentrator elements for plant

Project Results:
Note: In the text below, references are made to figures that can be found in the attached file named “figures”.

A) Design of fibre reinforced components for the parabolic trough
The initial parabola which had to be produced was 6m wide and 12m long. Restrictions with the manufacturing process led to adaptations in the parabola dimensions to lower dimensions (the initial reduction led to a parabola of 8.1m length and 5.5m width; the final dimensions of the prototype were 3.9m wide and 4.5m long).
The parabola was split it in 54 panels, 18 of which are Inner panels (red ones in figure 1), 18 are Middle panels (blue ones in figure 1) and 18 are Outer panels (green ones in figure 1). The prototype has 100 stiffeners (purple ones in figure 1) in order to join the panels together and provide stiffness to the structure. End plates and close hats have been designed to close the reflector of outside, preventing moisture absorption and adverse external conditions.
Each individual panel is produced using carbon fibre prepreg in a sandwich structure. The core of the structure is covered by Poly-Urethane (PU) foam, as can be seen in figure 2.
The stiffeners are carbon fibre laminates with a cross shape as can be seen in figure 3.
The end hat and end plates are also carbon fibre laminates. Their shape can be seen in figure 4.
In order to define, monitor and trace all these components in the overall structure a configuration management tool has been developed in Excel (the tool is attached to the report). The Configuration Management spreadsheet is a tool that establishes and maintains consistency in the assembly and the parts. It gives very specific information about the assembly and its parts, such as design, drawings, tooling and process control forms.
The connection of the panels is achieved by threaded bars that position and hold the panels in place. Figures 5 and 6 exhibit the location of the bars. This solution enables the easy access to each panel and facilitates easy removal and replacement of individual panels without having to dismantle the whole structure.

B) Manufacturing process for the production of the reinforced components
We have selected the vacuum bag process for the manufacturing of the panels and the stiffeners. The panels are sandwich structures composed by two carbon fibre laminates and a PU foam core. The flexible mirror is placed in the same cured process to save time and money. Therefore, the manufacturing lay-up would follow four steps:
Step 1: Place the flexible mirror onto the mould,
Step 2: Place the front carbon laminate,
Step 3: Place the PU foam core and
Step 4: Place the back carbon laminate.
A vacuum bag will be fixed around the sandwich panel and the air will be extracted with a vacuum pump. The application of pressure will consolidate the prepreg and bond the flexible mirror in the structure. The whole lay-up process can be seen in figure 7.
The layout for the manufacturing of the stiffeners can be seen in figure 8.

Taking as basis key dimensional parameters, such as maximum reflective surface available, distance between supporting poles or parabola’s geometry, the existing design can adapted almost instantly to offer the new panels distribution, with new dimensions for both panels and assembling elements (stiffeners). In this way, the design of the structure can be scaled up and down according to specifications with minimum effort (no need to re-design).

Due to the selected process for trough’s assembly, it is possible to extend the length of the parabola provided that we can consider longer distance between the poles that hold the structure together. The concept is presented in figure 9.

C) A novel highly reflective and protective coating for CSP
The main specifications for the coating are summarised below:
Develop a PVD (physical vapour deposition) coating that provides specular reflectance of >90% (300-2500nm), is corrosion resistant and UV stable. The coating must be applied directly onto the composite, and have a durability of >15 years under typical operating conditions.

The research was focused on:
i) adhesion and bond strength of PVD reflective coatings on e.g. Polycarbonate surface. Bond strength value of > 40 mN/m was the target.
ii) Determination of surface finish requirements in order to satisfy specular reflectance requirements
iii) Durability (>15 years) of a coating with a specular reflectance of ~94% that can be applied using PVD
iv) Cost: the application process should cost less than €40/m2

The resulting coating is described in figure 10. The coating has a SiO2 external layer for environmental protection, a second Alumina (Al2O3) layer that acts as an additional protective barrier, a 99.99% Ag layer for reflection of light and a layer of copper in order to facilitate stronger bonding with the polymeric substrate.

D) Coating process for application to the composite structure
Several possibilities were considered for the deposition of Ag on composite surfaces with different protective layers that should protect the Ag reflective coating from environmental conditions. Thin high transparent polymer foils and ultra-thin flexible glass substrates were also considered as protective sheets-layers of Ag reflective deposition towards the environment. Figure 11 shows indicative test samples and corresponding SEM analysis.
The direct deposition of the coating to the composite surface was attempted. The results (interfacial strength measured in pull-out tests) were not good due to the fact that the composite surface included both areas of polymer matrix and carbon fibres; two materials with very different adhesion properties. Figure 12 shows some of the attempts made.
An alternative process was successfully tried in order to overcome the problem. An intermediate layer was introduced between the composite and the coating. This layer allowed the successful coating of Ag. The materials tried for the intermediate layer were PET, PC and silica glass. Representative specimens are shown in figure 13.

E) Bonding layer between prepreg and aluminium layer
The Aluminium layer is adhesively bonded to each panel as part of the manufacturing process as can be seen in figure 7. A Process Control Form has been created for the case of an inner panel, ensuring the correct process of manufacturing. Thereby, all changes caused during the manufacturing process can be registered individually for each part.

The aluminium foil is supplied with two protective plastic plies. The first one, protecting the reflective side of the foil, must be kept in place as long as possible until the moment the trough is installed. The second one which keeps the back side of the foil has to be removed before the first ply of the prepreg is applied on top of it.

The tools for the process have been designed in order to meet the requirements of vacuum bag processes and have been modified to consider spring back effects.

The ‘spring back’ of the laminates is an undesired effect that results from the mismatch between the thermal expansion coefficients of the different plies in a composite non-flat laminate. It will depend on the material, the stacking sequence and the curing temperature.

Due to the difference in mechanical behaviour between a think non-structural glass ply and a thicker structural aluminium ply, the spring back should consider the presence of the aluminium. This implies two different set of moulding tools, one for glass reflective surface and the second one for aluminium surface.

F) Validated concentrator elements for plant
A collector system has been erected for the validation of the prototype part (Figure 14). A module length of 4m could be accommodated with an aperture width of 4.3m. This width allows connection to the existing collector and tracking system.
In order to reduce installation costs a novel founding system which includes anchor bolts instead of a concrete slabs has been foreseen. For connecting the pylon to the anchor bolts a plate is necessary. Figure 15 displays the sourced materials for the construction.


Potential Impact:
The final results of the project are:
• Successful deposition of Ag coating through PVD process to intermediate layer and bonding to the composite
• Detailed design of the composite part in a way that production can be scalable to larger structures
• The composite part comprises of smaller panels which are easier to manufacture, install, repair and replace
• Configuration management software for full traceability of each individual panel and each assembly

The potential impacts from COMPOSOL are:
• Manufacturing of composite troughs following modular design of the structure:
a) Easy manufacturing process since the composite parts are smaller in size and easier in geometry.
b) Easy to repair and replace faulty parts without having to replace the whole structure (cost reduction in case of damage)
• Manufacturing of composite parts using conventional vacuum bagging process: No need for expensive bespoke tooling, possibility to source the manufacturing to many SMEs with standard composite manufacturing capability (opening up of the supply chain, mitigation of the risk of having a small number of potential suppliers of the parts)