SOJ Materials Science & Engineering

This contribution presents alternative methods to cure thermoset composites in a more energy-efficient way. Therefore, three low energy demanding and fast curing sources which are UV, electron beam and infrared are selected as curing technology. For the UV and electron beam curing, special matrix formulations are developed, and these are combined with glass and carbon fibre reinforcements. For infrared curing, conventional thermoset epoxies are used and the curing parameters are optimized to prepare carbon and basalt composites. Finally, the flexural modulus and strength (3-point bending method), and glass transition temperatures (DSC) of the obtained composites samples are determined.

Fibre reinforced polymers (also called composites) already play an important role in our daily lives. These materials are lightweight and have high strengths, especially in the direction of the fibre. They find applications in buildings, automotive, aerospace, sports and consumer goods as lightweight alternatives to steel, aluminium and concrete. They are composed of reinforcing fibres (mostly glass, carbon, or basalt) and a matrix. The matrix can be thermosetting polymer (e.g. epoxy, vinylester, and polyester) or thermoplastic polymer(e.g. PP, PA, PLA). The curing step of the thermosetting matrix is often one of the most time and energy consuming steps in composite manufacturing. This leads to long processing times and slow production rates in combination with high energy costs.

An alternative possibility to crosslink or cure resins is the use of radiative energy instead of thermal curing. The advantage of radiation curing is that the energy is directly transferred to the resin while for conventional thermal curing heat is transferred by convection of the heated gases to the surface of material to be heated. Examples of radiation curing technologies are Ultra Violet (UV), Electron Beam (EB), Infrared (IR) and microwave curing[1]. UV curing is one of the most well-known methods and already frequently utilized in the wood and graphical sector. UV curing is a very fast process and can therefore significantly reduce production time. Additionally, UV curable formulations are either solvent free (100% systems) or water based, resulting in low VOC emissions[2]. UV curing systems also require less energy and space compared to large and high-energy-consuming ovens. This makes UV curing a sustainable and eco-friendly curing technology.

UV curable matrix formulations require adapted chemistries such as UV curable resins (mostly acrylate-type binders), viscosity modifiers (thickening agents for water based systems, viscosity reducers, i.e. monomers, for 100% systems) and photoinitiators. The resins can be epoxy, polyurethane or polyester based making them interesting composite matrix material[3].

However, a disadvantage of UV curing is the limited penetration depth of UV light. Material that is not exposed to UV light will not cure. This also means that UV curing is nowadays mainly used for repairing purposes in the composite industry or for filament winding processes[4,5].

Fortunately, there are methods to increase the curing depth, namely 1) selecting UV transparent fibres, 2) using through-cure photoinitiators and 3) using UVA radiation. The combination of these 3 factors creates opportunities to cure relatively thick composite panels by UV radiation.

Specifically, UVA-LED lamps are very interesting. Unlike conventional UV curing, UVA-LED systems only emit narrowband UVA light meaning that no toxic ozone is formed and no thermal stress from IR radiation occurs[6]. Additionally, UVA is the most deeply penetrative of the UV light range, allowing for curing of thicker composites.

For EB curing, the same chemistry can be used as for UV curing, but there is no need to add a photoinitiator. The electrons generated by the EB source already have a sufficiently high energy to radicalize and polymerize the oligomers and monomers. Similar to UV curing, EB consumes less energy and cures formulations in seconds or minutes instead of the hours or days typical for thermal curing of composites. Unlike UV curing, no transparency of fibres and matrix is required, but penetration depth is a function of material density and acceleration voltage. The lower the density of the fibre reinforcement the better the through-curing. This means that carbon is a more suitable fibre than glass or basalt fibre as the density of carbon fibre is around 1.8 g/cc while for glass and basalt, density is around 2.6 g/cc[7].

When relying on the more conventional thermoset epoxy resins, IR curing can be an interesting possibility. When infrared radiation hits a material, molecular oscillations or vibrations take place, causing a heating effect within the target material. Like light, infrared radiation does not require a medium, thus it can be transmitted through a vacuum. Infrared heating can offer a faster heating time, reducing the oven length and increasing line speed when compared with traditional convective heating ovens. In a study by Kumar et al.[8], it was shown that IR curing reduced the curing time by a factor 4 compared to conventional curing in an oven. Besides full curing, infrared can also assist fibre drying, debulking (reducing the void content in the uncured composite) and pre-curing (i.e. preparation of prepregs).

Microwave curing is also suitable for the curing of conventional epoxy resins. Heating is based on the dipolar molecular interation with the electromagnetic field. The heating efficiency is strongly dependent on the dielectric properties of the materials making it more suitable for glass fibres than carbon fibres. However, adapting the electromagnetic frequency and heating field makes it possible to cure carbon reinforced materials[1].

In this article, UV-LED, EB and infra red curing were used to cure fibre reinforced polymers efficiently. For UV-LED and EB special matrix formulations based on polymeric and monomeric acrylates were developed. Glass fabrics were selected as the fibre reinforcement for UV-LED curing and carbon fibre reinforcements for EB curing. For IR curing, conventional thermoset 2 component (2K) curing epoxies were selected and combined with basalt and carbon fibre reinforcements. For each technology, good processing parameters were determined, and the flexural properties and glass transition temperatures of the obtained composites were determined and compared with a conventional cured glass/epoxy composite.

Glass twill fabrics (390 g/m²) and carbon twill fabrics were bought from MC Technics. Basalt twill (220 g/m²) and atlas (350 g/m²) fabrics were bought from Basaltex.

The chemicals were supplied by BASF, Allnex and Sicomin. An air-cooled LED lamp from Sadechaf was used for the UV curing and the wavelength of the lamp is 395 nm and the maximum power is 3.2 W/cm². EB curing was performed at Ebeam Technologies using an electron beam lamp of 5.1 kW and acceleration voltage of 300 kV. IR curing was done with an IR lamp from Heraus with a maximum power of 3.5kW.

Synthesis
UV-LED cured composites were prepared by first mixing an epoxy acrylate oligomer diluted with a triacrylate monomer (30%) and a photo initiator wich absorbs UV light at 397 nm. This formulation was used to impregnate the glass fabrics. Eight layers of impregnated glass fabrics were stacked and cured for 30s with the LED lamp at ambient atmosphere (hand lay-up) and under vacuum (vacuum bagging; see figure 1).

EB cured composites were prepared by first impregnating carbon fabrics with a biobased aliphatic diacrylate diluted with a triacrylate monomer (18%) followed by EB curing with an acceleration voltage of 300 kV. Different stacks were prepared in order to find the maximum curing depth. An overview is given in table 1.

IR cured composites were prepared by first impregnating basalt and carbon fabrics with a 2K epoxy resin which needs - according to the TDS - 90 minutes curing at 120°C to obtain a glass transition temperature of 107°C. Curing occurred by IR in ambient atmosphere and under vacuum at different curing times. An overview is presented in table 2:

Characterization
To determine the flexural modulus and strength of the UVLED cured matrices and composites, a 3-point bending test (ISO 14125) was performed. For each composition, 5 samples were measured. The glass transition temperature was determined by a differential scanning calorimeter (DSC) from TA Instruments. Scanning Electron Microscopy (SEM) images were obtained by a JSM-7600F FEG-SEM from JEOL with a resolution of 2 nm at 2 kV and under high vacuum.The samples were precoated with a 4nm thick platinum/palladium layer via sputtering to minimize electrical charging of the samples.

For the preparation of UV-LED cured glass fibres reinforced composites, the matrix concentration was varied from 25% to 45% in steps of 5%. For the hand lay-up, 25% was not enough because the glass fiber layers delaminate after curing, while 45% was too much (resin leaches out). For the vacuum bagging 40%

was already too much. The thickness of the cured samples was measured, and the results are summarized in table 3.

As could be expected vacuum bagging leads to thinner samples. Also, it was no problem to cure composite plates with thicknesses of 3 mm, although UV curing is almost only known for the curing of thin coatings (μm range). Furthermore, LED light is especially suitable for bagging as the ‘cold’ light will not excessively heat up the thermoplastic vacuum bag, which would be the case if conventional UV curing were used. Another advantage of the vacuum bagging method is that there is no risk of oxygen inhibition which would hinder the surface curing.

The flexural modulus and strength of the composite samples were determined and compared with a room temperature cured 2K epoxy composite prepared by hand lay-up.

The results are presented in figures 2 & 3.Vacuum bagging results in more uniform composites shown by the reduced error when compared to the hand lay-up method. The flexural modulus is also higher and increases by decreasing the matrix concentration (from 15 to 17 GPa). The influence on the flexural strength (figure 3) is less pronounced.

For the UV-LED cured hand lay-up samples, higher matrix loadings lead to better mechanical properties. The best mechanical properties are obtained for the samples that contain 35 or 40% matrix with values around 14 GPa for the modulus and 200-240 MPa for the strength. The reason why the values for modulus and strength of the sample that contains 30% matrix are lower is because 30% is not enough to hold the glass fabrics together when exposed to mechanical pressure. This results in a faster delamination and thus a less strong composite.

When looking at the 2K epoxy composite it can be seen that the strength is about 50 MPa higher, but the modulus is significantly lower (10 GPa) than the UV-LED cured samples at 35% matrix concentration.

The glass transition temperature of the UV-LED cured composites was found to be 58°C which is comparable with the Tg of the 2K epoxy (55°C). Furthermore, the UV curable formulations have a long working time (several hours when working in artificial light and ready to use formulations can be stored for more than a year when stored in the dark) and a very short curing time (seconds to a few minutes) with an energy consumption of about 0.27kWh per m², while the 2K epoxy resin has a limited processing time (+/- 30 minutes) and a long curing time. 24 h is needed for sufficient curing and 7 days are needed for complete curing. This means that UV-LED curing is an interesting alternative for making glass fibre reinforced composite structures.

The curing experiments revealed that when more than 4 layers were stacked, the matrix in the bulk of the composite was not cured. This means that the maximum curing depth is about 1mm for the carbon composites. The time needed to cure the composites (A4 size) is 12s with an energy consumption of 0.27 kWh for 1m² which is similar to UV-LED curing. The Tg value was found to be 135°C making it suitable to utilise this composite at elevated temperatures. The flexural properties of the composite were determined; the flexural modulus is 33 GPa and the strength is 189 MPa. These values are quite low considering that LED cured glass fibre composites already have strengths higher than 200 MPa. To find an explanation of these low mechanical properties, the cross section of the composite was analyzed with SEM (figure) The image reveals a high porosity and a low fibre/ matrix interaction resulting in a low-quality composite material.

The processing conditions, flexural properties and glass transition temperatures of basalt IR cured composites are presented in table 4.

From the table, it can be observed that the glass transition temperatures are comparable and also only a little bit lower than the Tg mentioned in the datasheet of the resin (107°C) showing that IR technology is suitable for the curing of epoxy based composites. Further curing under vacuum results in higher mechanical properties. However, the properties are still lower than expected. Therefore, SEM imaging was used to characterise the cross section of the vacuum bagged composite and is shown in figure 5.

The SEM image of the cross section reveals that the sample still exhibits a high amount of porosity which explains the lower mechanical properties. To reduce the porosity of the samples a new curing process was developed and basalt (IR3) and carbon (IR4) composites were prepared. The curing time was increased from 42 to 90 min to allow a better spreading of the resin and a breather cloth was introduced to avoid air being trapped in the composite material. Despite initial concerns that a breather fabric would inhibit the IR radiation and prevent the resin curing, the curing process was not negatively impacted.

The energy consumption during the curing was also monitored this time and the normalized total energy consumption to cure a

composite with thickness of 2.5 mm is 17 kWh/m². Compared to UV curing this consumption is much higher, but still significantly lower than thermal curing which was found to be 26 kWh/m² for an electrical furnace comparable in size of the IR equipment. Further, a better design of the IR setup could lead to an even lower energy consumption, showing that IR is a more energy efficient curing technique than conventional curing in a furnace.

With the addition of the breather cloth the flexural properties are increased leading to a modulus of 23 GPa and strength of 903 MPa for the basalt composite and a modulus of 34 GPa and strength of 634 MPa for the carbon composite.

When looking at the SEM pictures (figures 6&7) of composite IR3 and IR4 no porosity is observed, as expected, given the high quality of the samples.

Further, as expected, the stiffest materials are obtained with carbon fibre. However, normally the strength of the carbon fibre should be higher than the basalt fibre composite as carbon fibres are stronger fibres. Calculations of the fibre volume fraction reveal that these values were lower for the carbon twill (0.33), than for the basalt fibre composite sample (0.46) which explains the lower strength of the carbon sample.

It is clear from the results that the tested radiation technologies are very suitable for the curing of composites. UV curing and EB curing have the advantage that VOC free & low viscosity formulations can be used with a very long potlife and that fast processing speeds with low energy consumption are possible. The mechanical properties are lower than the composites cured by IR curing. It needs to be taken into account that UV curing only occurs when the matrix is exposed to UV light. This makes this technology especially suitable for fibre reinforcements of glass fibres. Their mechanical properties, that are lower than carbon fibres, can be improved by curing under vacuum. Typical vacuum bagging accessories which improve the quality of the composite such as peel plies and breather cloths cannot be placed on top of the material due to their UV blocking effect, making it difficult to obtain high quality composite materials. As not all composites need to achieve those very high mechanical properties and a lot of them are still prepared by hand lay-up, UV cured composites can be a good alternative for glass/polyester based composites. The advantage of EB curing is that high tech carbon fibres can be used for composite manufacturing and similar fast curing speeds are achieved as with UV curing. The 300 accelaration voltage used in our tests limits the curing depth (about 500 μm in one shot). There do exist Ebeams with acceleration voltages of 1MV, but they are quite expensive and and thus not always economically feasible. So, EB is to be used for thin laminates.It was shown that the mechanical properties are strongly affected by the high porosity and low fibre-matrix interaction of the laminates. Developing a vacuum assisted (curing/infusion) curing set up and a proper sizing of the fibres will definitely improve the mechanical properties of the composites. IR curing results in composites that have similar high properties as conventional cured composites. The curing time and temperature of the resin must be respected to avoid over and undercuring of the resin especially for thicker composite parts meaning that gain in curing and time efficiency is mostly obtained by the ability to heat the resin more quickly to the desired temperature. With this approach it is shown that a 34% reduction in energy can be achieved.

The results of this research show that the three alternative curing technologies have the potential to augment conventional curing methods. The pros and cons of each technology were discussed. UV (LED) curing is a fast and energy efficient curing method, suitable for glass fibres and a good alternative for conventional hand lay-up made composites. EB curing is also a fast and energy efficient curing method, most suitable for thin sheets of carbon fibres. IR curing is an energy efficient method to cure conventional thermal curing epoxy resins. The most promising application is to replace the thermal curing step in vacuum bagging processes.

The IGF research project 193 EBR (FENECOM) is initiated by the research association Forschungskuratorium Textil e.V., Reinhardtstraße 12-14, 10117 Berlin.

The Fenecom project has been funded by the AiF within the programme for sponsorship by Industrial Joint Research (IGF) of the German Federal Ministry of Economic Affairs and Energy based on an enactment of the German Parliament and has been funded by Flanders Innovation & Entrepreneurship (VLAIO HBC.20160666)

1. DilmuratAbliza, YugangDuana, Leif Steuernagel, Lei Xiea, DichenLia, Gerhard Ziegmannb, Curing Methods for Advanced Polymer Composites - A Review; Polymers & Polymer Composites. 2013; 21:341-348.
2. E Kiyoi. The state of UV-LED curing: an investigation of chemistry and applications, Radtech report. 2013;2:32-36.
3. Frederik Goethals, MyriamVanneste, Ralf Lungwitz, Martina Janickova, Jan Marek. Eco-efficiently cured and halogen-free fire retardant composites, Proceedings ICCM20 conference. track 1-03. 2015.
4. Guoqiang Li, Pourmohamadian N, Cygan A, Peck J, Helms JE, Pang S. Fast repair of laminated beams using UV curing composites. Composite Structures. 2003;60:73-81.
5. Susan Mitchell.https://www.slideshare.net/SusanMitchell8/uv-curing-cure-on-demand-for-filament-winding
6. Chad Taggard, Sara Jennings.
7. https://www.ebeam.com/news-archive/archive/2016/eb-vs-uv-penetrationas
8. P Kiran Kumar, NV Raghavendra, BK Sridhara. Development of infrared radiation curing system for fiber reinforced polymer composites: An experimental investigation. Indian Journal of Engineering & Material Sciences. 2011;18:24-30.