Wind blades circularity - Resins development for improved sustainability

Wind turbines play a vital role in the transition to greener energy and to fight climate change. In the coming years the number of installed turbines will increase dramatically. Consequently, the sustainability and especially the circularity of wind turbine components becomes more and more important. Many components of a wind turbine are made from recyclable materials like steel or aluminum, but blades are made from composite and are with that the biggest non-recyclable structure of a wind turbine. The resin is key for the circularity of wind blades. Glass and carbon fibers offer the strength and stiffness to build long wind blades, core materials like balsa and polymeric foams are applied to reduce weight, and the resin secures fiber straightness, transfers stresses and glues everything together. If it is possible to remove the resin in a mild and efficient way, the other materials can be recovered and reused. This is the reason why the research in the field of composites recycling has intensified over the last years. Two main directions have been followed in the last years: Discovering a recycling method for blades and in general composite structures based on conventional resin technology or implementation of chemically modified or tailored resins for recyclability by design


Wind blades and recycling
Wind turbines are generally perceived as green because they are used for green energy production.But at the same time the public interest in sustainability of wind turbines is growing and several aspects of sustainability and pollution are questioned.
Blades are huge parts, very visible, and made from materials difficult to recycle.But recycling is just one part of the waste management and not the one with the highest impact.As shown in the waste hierarchy established in the waste framework directive [1] (Figure 1) the avoidance of waste is highest priority.The material that is not used has the lowest environmental impact.Despite the amount of waste, for wind blades landfill has been the dominant end-of-life option.With the emerging landfill bans other recycling strategies must be identified, and cement co-processing is at the highest readiness level.Pyrolysis is an emerging technology to recover fibers, the resin is decomposed and recovered as an oily residue.
Solvolysis can range from mild conditions using benign chemicals to supercritical conditions and nasty chemicals.The broad range of process conditions is also exhibited in the condition of the recovered fibers and core materials.
A different approach is the manufacture of noise barrier panels for acoustic applications , such as the commercial products offered by Miljøskᴁrm [2].The composite parts of the blade are converted in to granules, which are then used to manufacture the panels.
In terms of re-use, the application of blades or blade parts for architectural purposes has been shown.Examples are bike sheds from parts of the blade shell, blades as load bearing elements of bridges or blade shell pieces as sunscreens for buildings [3].

Evaluation of recycling methods
With an increasing number of recycling methods available, it becomes more important to be able to compare the recycling methods based on objective measures.Different measures can be applied and will give a broad and good insight into the characteristics, strengths, and weaknesses of different approaches.Some examples of how to measure and compare recycling methods are: Cost value proposition to evaluate process cost and value of recovered materials, CO2 footprint of the process and CO2 saving using the recovered materials [4], [5].

Solvolysis of conventional resin systems
The solvolysis of resin systems for wind blades is of high interest.But Solvolysis conflicts with the fact that epoxy amine resin systems have been designed to be strong and chemically resistant.Nevertheless, some interesting processes have been presented over the last few years.
It was shown that a recycling process discovered by Kaufmann et al. at TU Clausthal and MPM Environment Intelligence GmbH.based on boron halides is able to remove the epoxy resin completely.The process is compatible with glass fibers as well as carbon fibers [6].
Another example is the process developed by Catack-H in Korea focussing on the recycling of carbon fiber-based composites.First the decommissioned part is pre-conditioned in a liquid for 24 h, then the actual recycling process takes place [7].
Lately also a two-step recycling process developed in the CETEC project was presented [8].The first step is performed to separate the resin from the other materials, which can be recovered and used thereafter.The second step is conducted to valorize the remains of the resin coming out of the first step.So far, no details on the two process steps have been disclosed, so it is not possible to judge the feasibility or compare with other processes.

Pyrolysis of conventional resin systems.
Pyrolysis intends the separation and recovery of composite materials by thermally decomposing the matrix and recovering the fibers afterwards, due to the higher temperature resistance of carbon and glass reinforcements compared to the cured resins.Alternatively, pyrolysis can be used only to extract energy with the combustion process, sometimes using it in closed loop, or it could be used to recover fibers and degraded matrix in the form of oil.Currently being used at industrial scale for other applications, pyrolysis for blade materials is still at a low technology readiness level.
Pyrolysis has been shown [9] to recover carbon fibers with minimal presence of pyrolytic carbon (Figure 2), organic residues and oxidation, thus retaining a significant level of mechanical properties.The recovery of carbon fibers in these conditions depends mostly on oven atmosphere, temperature and heating rates.When recovering GFRP, the cost balance of the process is much less beneficial, given the higher extent of damage inflicted to the glass fiber surface, lowering the mechanical properties and the lower cost of the glass per kg, which do not justify the large investment and running cost pyrolysis commonly involve.To overcome this limitation, microwave pyrolysis is being investigated [10] to enable faster energy transfer an shorter processing times, which both decreases damage to the fibers and the overall process cost.
Fluidized Bed Reactors are also investigated for blade composites recycling, due to its multi-material versatility and lower process cost.At the moment, FBR systems can only work fed with grinded blade composite residues, thus adding the limitation of only recovering short fibers to the already existing one of degraded properties.

Recyclability by design of resins
Recyclability by design with respect to resins means the introduction of chemical groups or additives into the resin.By chemical modification it is possible to enhance the recyclability of resins.The chemical modification can happen in different ways: 1) Additives can be used which are not integrated into the polymer network.
2) Functional groups can be introduced into the polymeric network.Functional groups that are integrated in the polymer network can either be randomly distributed or located at defined positions in the network.The resin chemistry and especially the mechanism of polymerization is a deciding factor of how controlled the introduction of new functional groups into the polymer network can happen.Polyester and vinylester resins form the network in a chain growth reaction mechanism.Di-or multifunctional oligomers or polymers that contain unsaturated bonds in the main chain as side chains are dissolved in styrene and divinylbenzene or (meth-)acrylic monomers.The reaction is initiated by radical forming initiators (eg.peroxides) and continues in a random fashion by chain extension via reaction of the radicals with double bonds and transfer of the radical to the new chain end.This complicates the directed introduction of functional groups.
Special initiator or catalyst systems allow more controlled polymerization.Examples are RAFT or ATRP polymerization of polyester and vinylester resins.Application of these and similar methods offering living polymerization allows more controlled chain and network formation, and the creation of fully latent resins [11].
Epoxy-amine infusion resins react in a step-growth mechanism.The reaction starts simultaneously in the whole bulk resin.The difunctional epoxy resin reacts with the tetrafunctional amine hardener.With this choice of components the crosslinking happens only through the hardener.As a consequence the modification of the hardener backbone can introduce functional groups into specific and directed locations of the network.

Introduction of cleavable moieties
The introduction of cleavable moieties has been proven in literature and commercial scale.A good example is Aditya Birla Chemicals´ Briozen product range.The technology builds on the introduction of acetal or ketal groups into the hardener [12] The amine functional hardener can then replace conventional amine hardeners in the formulation of epoxy-amine resin systems.The strength of the technology is, that under neutral conditions the resin is stable and offers the same performance as resins using conventional amine hardeners.First in contact with organic acids, prefentially acetic acid, in aqueous solution and elevated temperature, the cleavage of the acetal or ketal group is catalyzed forming dissolvable polymer chains out of the thermoset network originally formed.
Another example is the recently presented EzCiclo resin technology by Swancor [13].The epoxy part is modified to introduce a cleavable moiety.In this case the hardener is no different to conventional resins.The recycling process requires a proprietary recycling solution called CleaVer.The composite based on EzCiclo resin is immersed in the CleaVer liquid and heated to 120-140 °C.The cleavable bonds are opened and the resin is dissolved in the Cleaver.The reinforcement fibers can be collected and reused after a cleaning step, the Cleaver solution containing the recyclate can be used as hardener for epoxy resins for example in flooring applications.

Introduction of reversible bonds
The introduction of reversible bonds into the network is a similar strategy to enhance the recyclability of the resin.But the introduction of reversible bonds adds further features to the resin and the resulting composite.The controlled opening and closing of bonds in the network allow the regulation of the crosslink density of the resin.With a smart positioning of the reversible bonds, the crosslink density of a thermoset polymer network can be reduced and in the most extreme case shifted to a thermoplastic linear polymer, triggered by an external stimulus.The opening and formation of bonds is a dynamic process and happens simultaneously in an equilibrium state.The bonds are not necessarily reformed at the same position.This results in viscous behavior and enables operations that are normally limited to thermoplastic materials.
A known example for reversible bonds is the Diels-Alder reaction.It is a versatile reaction that forms a six-membered ring from a molecule containing two conjugated double bonds (diene) and another molecule or chain of the same molecule containing one double bond (ene).The reaction is shown in Figure 3 on the example of the dimerization of cyclopentadiene.[14] At room temperature the equilibrium lies at the 6-membered ring, at increased temperature the equilibrium switches to the open state of diene and ene.Although the concept is well known [15], the application for infusion resins is difficult because the integration of the functional groups into resin and/or hardener result in bulky molecules having a negative impact on the viscosity.

Introduction of dynamic bonds
A closely related technology to the introduction of reversible bonds is the application of dynamic bonds, the concept is called covalent adaptable network.Two different mechanisms are described in literature: Dissociative covalent adaptable networks, where upon an external trigger, which is normally heat, bonds are broken and then reformed.
Several chemical concepts have been investigated and presented over the last few years.Two prominent examples are resins based on disulfide bond exchange (Figure 4) [16] and resins based on imine exchange (Figure 5) [17].Both resin technologies behave like conventional thermoset resins until a certain temperature is reached.Above this temperature the active bonds start to rearrange, but the crosslink density is constant.The dynamic bond exchange results in viscous flow, which is enough to enable options, that are normally limited to thermoplastic resins.The extraordinary behavior also opens up for new recycling techniques.Tailormade solvolysis processes can be applied to conduct a depolymerization of the network under mild conditions.Given the similar quality of the fibers recovered, the process cost is the determining factor for the fiber price.Process cost is controlled by factors like process temperature and time of the recycling process, cost of chemicals used and simplicity of the process.Similar to the waste hierarchy a classification of recycling processes for polymers has been proposed by Zero Waste Europe (Figure 6) [18] with respect to the recovered material.
Figure 6: Classification of polymer recycling processes [18] The solvolysis process degrades the thermoset network of the composite matrix.Depending on the underlying technology the cleavage of the network results in different products.To compare the technologies one aspect is the conservation of functionality and molecular structure, as shown above, and the atom economy.Energy and resources have been consumed in the synthesis of resins and its curing.The more the structure can be conserved and integrated into a valuable product the better.
A cradle-to-cradle recycling process is not automatically the preferred solution over a downcycling process, if the cradle-to-cradle recycling process goes back several steps in the synthetic pathway of the resin components and requires steps of re-synthesis to yield these components again.
The atom economy concept can be adjusted for chemical recycling.The functionality, complexity and purity of the recovered components can be taken into account.The more of the functionality and complexity is recovered at the highest purity, the better.By-products should be avoided as good as possible.
As an example an evaluation of the Aditya Birla Chemicals Briozen resin based on Recyclamine® technology shows, that although the cured resin is converted from a crosslinked thermoset to a thermoplastic resin, almost the full functionality of the resin system is maintained and the thermoplastic resin can be recovered in high purity.Only a small molecule is cleaved off in the recycling process and can be recovered and reused as well.
Another example for comparison is a recycling process, that is able to recover 50 % of the resin system as raw material for the resynthesis of a resin component.The other 50 % are a mixture of different components and can only be used for energy recovery.
Although the latter process offers real cradle-to-cradle recycling for a part of the cured resin, only 50 % of the resin is actually recycled, a poor atom economy.Also the functionality and complexity are reduced coming from the cured resin to the raw material for a resin component.In this comparison of the two technologies, the downcycling appears as the preferred solution over the cradle-to-cradle approach.

Performance of recyclable resins on basis of Aditya Birla Chemicals´ Briozen resin
The SGRE RecyclableBlade has been a ground-breaking development towards recyclable blades.The RecyclableBlade has been introduced using Aditya Birla Chemicals Briozen resins which are based on the Recyclamine® technology.
The concept of the Recyclamine® technology allows the synthesis of molecules with defined molecular structure and molecular weight.The components offer similar structural features as the commonly used hardener components.
Acetal groups are already long known in polymer chemistry and find application in polyoxymethylene (POM) or polyvinylbutyrate (PVB) to name some examples.
The Aditya Birla Chemicals Briozen infusion resin is a good example for the smart modification of traditional epoxy amine resins for enhanced recyclability.The general type of resin is still epoxy amine, but the hardener comes with a cleavable group in the backbone.The technology allows the formulation of resins with similar mechanical and process performance.To find suitable resins several different hardeners have been developed by Aditya Birla Chemicals.The hardeners offer different pot life and curing time to have the best fitting solution for each application.Also, the weatherability is not negatively impacted.

Mechanical properties and interface adhesion
SGRE produces blades in two different ways: IntegralBlades® that are manufactured in a one-shot infusion.So-called "butterfly blades" are manufactured by infusion in several pieces, that are later assembled in a bonding step.The individual pieces of the butterfly blades are smaller in size than the shell of an IntegralBlade®.The infusion is faster, so the pot life can be shorter than for the infusion of an IntegralBlade®.
To have a fitting product for both processes dedicated hardener systems have been developed by Aditya Birla Chemicals which are then combined with the same resin base.Both resin types, the one for IntegralBlades® and butterfly blades, have been thoroughly tested and characterized.The test results showed that throughout the tests the performance is equal to better compared to conventional resins systems.This has been observed for neat resin tests (Figure 7), static (Figure 8) and fatigue tests on laminates level as well as interface tests to hand-lamination repair resins and bonding paste (Figure 9).

Weathering stability of the cured resin
Because the blades are exposed to the environment for their whole operational lifetime, the weathering resistance of the resin has been investigated.On one hand the resistance to different acidic conditions was tested in full immersion tests following ISO175.The weight change before (Wiped dry with a cotton cloth) and after drying was checked.The summary of the results is shown in Figure 10.The full immersion tests show little weight change and generally the same trend of the Briozen resin compared to conventional resin used in blades production.Only the 5 % acetic acid solution shows the potential of acetic acid to swell the epoxy network by the weight change of the reference system.No significant effect of the pH, tested with three different buffer solutions, was observed.Acidic rain is an often-mentioned concern.Acidic rain is a mix of various acids typically composed of sulfur-and nitrogen-based acids and a pH of around 4. To mimic these conditions nitric acid and sulfuric acid were diluted to reach pH 4. Both the Briozen resin and the conventional resin did not exhibit any negative effect upon exposure.
Because the chemical nature of the resin is very similar to conventional DGEBA based resins, the UV stability is assumed to be as low as for conventional resin systems.This was validated as part of a weathering test campaign following ISO20340 modified to fit for composites.The test cycle is shown in Figure 11.Both resins show slight decrease of the weight during the test, and a non-significant difference of weight change after 25 weeks.Two effects are counteracting: Weight increase by water uptake and weight loss because of degradation under the test conditions.The recyclable resin shows less weight increase in the beginning and slightly more weight loss towards the end.This can indicate a higher water uptake of the reference resin compared to Recyclamine® resin.

Selectivity of the recycling process
The recycling process is efficient, but selective to the choice of acid and requires elevated temperatures.Representative acids and different conditions have been tested.The results are presented in Table 1 and Table 2.The above shown results demonstrate the selectivity of the recycling process.A strong, but inorganic acid with chloric acid as model substance was tested at 10 wt % concentration and 60 °C.At the same temperature and half the time, the mass loss with acetic acid was more than four times more.This indicates that low pH -this equals a high concentration of H3O + which is the catalytic species for the acetal cleavage -is not enough to cleave the acetal/ketal bonds of the Briozen resin.The same effect is observed with citric acid, which is not able to catalyze the recycling process.So neither inorganic acids nor a standard organic acid are able to catalyze the recycling process efficiently, the addition of DMSO, a good solvent for many organic substances, appears to have no positive effect.Acetic acid seems to offer another special feature other than just generating H3O + .A potential explanation can be the property of acetic acid to be able to swell epoxy resins.The excellent swelling behavior is most likely based on the ability to break hydrogen bonds and the character as good solvent.
The effect of the recycling process on other materials was investigated in a series of tests.Sandwich panels based on balsa and PET core were manufactured, composites containing carbon pultruded profiles were prepared.The machined specimens from the panels were immersed in the acetic acid solution (30 wt%) at 80 C. When the whole resin was dissolved from the glass laminate, the core material or carbon profiles was isolated, dried, and weighed.As reference a panel with conventional resin and balsa wood was tested under the same conditions.With the estimated mass loss of the machining the weight change was calculated.The results are presented in Table 3.The results of the balsa wood from the recyclable sandwich laminates show a significant weight increase.A potential explanation is the difficulty to completely remove resin which was taken up by the balsa wood during the infusion process.Also a significant change of the geometry because of swelling was observed.
PET foam was visually unaffected, neither the appearance nor the geometry changed during the recycling process.The weight change was not significant either.This can be explained by the much lower resin uptake of the closed cell structure of PET foam.
The carbon profiles appeared unaffected by the recycling process, but were recovered as one block, the recycling liquid seemed not to be able to penetrate well enough in between the profiles for dissolution of the resin and to allow for the separation of the individual carbon profiles.
The reference sandwich laminate shows a weight change of 12 % of the whole laminate, which is a combination of swelling of the resin and the balsa wood.The reference laminate was still intact, although in some cases the balsa deformed resulting in some delamination.
In a larger test on a representative coated sandwich composite specimen the recycling of a coated structure was investigated.The effect of the recycling process on the blade coating system was observed to be low.Interestingly the coating seems to be susceptible to the diluted acetic acid.At the process temperature the coating starts to blister, and the dissolution of the underlying resin appears to proceed almost unaffected.The coating comes off as a film.This is a good indication, showing that the preparation of the materials for the recycling process does not require a step removing the coating.

Conclusion
Recyclability is an important topic in wind industry and many related activities are currently running.Pyrolysis offers interesting potentials, but lately solvolysis has become a strong focus of investigations as well.Nevertheless, solvolysis of conventional resin systems is difficult and requires a sophisticated recycling process.
Resins that are specifically designed facilitate an easier recycling process.The cleavable bonds can be introduced in two different ways: randomly distributed or at defined locations.This influences the product of the recycling process together with the number of cleavable bonds.
A smart choice of the underlying technology allows to manufacture wind blades, that offer the same performance in all aspects as conventional blades, but with enhanced recyclability in a mild solvolysis process.The recovered materials are of good quality and high value, this results in an overall positive business case together with a relevant CO2 eq.saving potential.
The presented results indicate that recyclability by design of resins is a promising approach and has the potential to replace conventional systems.

43rd 6 4. 4 .
Risoe International Symposium on Materials Science IOP Conf.Series: Materials Science and Engineering 1293 (2023) 012006 IOP Publishing doi:10.1088/1757-899X/1293/1/012006Evaluation and comparison of solvolysis process of conventional resins and chemical concepts for enhanced recyclingLike for recycling methods overall it is beneficial to identify measures to evaluate and compare different concepts and technologies.As for the general comparison of recycling methods, the CO2 footprint and the cost value proposition are important measures.The recycling process conditions are similar resulting in the value of the recovered fibers being almost the same.Differences can be found in the core materials -in which quality is balsa, PET and other core materials recovered.

Figure 9 :
Figure 9: Interlaminar shear strength of Recyclamine® and reference resin laminate to laminate based on conventional epoxyamine repair resin and single lap shear strength of Recyclamine® and reference resin bonded with structural adhesive (bonding paste)

Figure 10 :
Figure 10: Weight change of Recyclamine® and reference resin based laminate samples after 24 h and 168 h full immersion in different test liquids.

Figure 11 :
Figure 11: Weathering test cycles following ISO20340 The uncoated laminates were tested for weight change and TG change with the test cycle of three days UV exposure and condensation, three days of salt-spray and one day at -20 C. The results are shown in Figure 12 and Figure 13.

Table 1 :
Test of different recycling conditions by weight loss of neat resin samples

Table 2 :
Test of different recycling conditions by weight loss of a thin biax glass fabric based laminate

Table 3 :
Mass change of materials subjected to recycling conditions