Carbon fiber reinforced composites are already established in niches in automotive manufacturing today. Especially aerospace cannot live without the lightweight advantages of this class of material. However, everyone observing the composite industry is waiting for the next development step, which is the breakthrough in scaling up to high-volume production. So the first carbon composites “hype” is actually over. In order to drastically reduce manufacturing costs the composite industry has to take a step back and look at the most significant challenges that come with mass production.
Read the article offering solutions to the current technological challenges:
a) composite part design
b) material properties and
c) composite processing itself
It all changed when BMW introduced the i3 electric car with a full CFRP body into serial production in 2013. Suddenly the majority of material researchers discovered and discussed its potential for high-volume production. Carbon fiber composites continue to be in trend today: they have excellent properties for lightweight construction and are present in the aerospace industry and in sports and leisure for a while now. Furthermore, these composites currently make their way into civil engineering and building and construction as well. So everything seems to go well and the market should be soaring, correct?
Well it isn’t – at least that’s what is obvious if you have been in close contact to the manufacturing industry for the last few years. The CFRP trend comes with a drawback that we still cannot see CFRP going into really high-volume serial production e.g. in automotive. And to cite an aircraft engineer I met recently at a trade fair: “If we are going to build all the airplanes in our order books like we do today – we will be broke!” The major challenge for carbon fiber composites over others is still price: price per kilogram of CFRP material is still a factor of 4 too expensive for automotive mass production.
It worked extremely well for the microchip industry – why shouldn’t it be possible for composite technology as well?
We at NETZSCH Analyzing & Testing are in the fascinating position to observe many of the research and development topics first hand as engineers use our thermal analysis instrumentation along the whole workflow of the carbon fiber composite production. In this way we are able to form our own view of what’s going on in the composite market and what next steps are necessary to overcome the current challenges.
Let’s start from scratch, looking at the things from the very beginning – let’s start with part design. The design of carbon fiber reinforced composites is very different from e.g. metal design as commonly taught at schools in mechanical engineering: metal for once is isotropic, CFRP is anisotropic with the highest strength in fiber direction. That changes the way that forces are being distributed in your material entirely. Furthermore, carbon fibers don’t form naturally, and we need an adhesive resin holding the fibers together and a sizing for a good interaction between fiber and resin.
But let’s go back to the main industries where composites are being used, aerospace and automotive: if you stick to the anisotropy principle, an absolute design no-go to get maximum performance would therefore be to drill holes into a CF structure and weaken the network in fiber direction. However – if you look at automotive and aircraft design, every so often designers cannot start from scratch but with a certain structure or part they want to replace with lightweight-optimized CFRP. What happens? The result might turn out to be a “good enough” design with respect to where you came from and make your car a bit lighter and a bit stiffer.
That’s for example what took Boeing and Airbus recently to “redefine the wing”1 from scratch taking into account all the unique properties that composites have. It will be highly interesting to learn what will come from this.
As the design is the very first step for a composite part, this makes it the most crucial step forward for CFRP high-volume production. And as the article about aerospace wing design shows, the industry is now starting to take the effort to go back to the very beginning of composite design and start over again to get the best possible product. We believe this is a step into the right direction.
CFRP composites are not just one material but you at least have to understand the material properties of the carbon fibers, the matrix resin and the interaction between both. So the second challenge might seem quite straightforward: in order to get your design right, you need to know the materials you use and how they interact to make the CFRP part. Furthermore, composites using thermosetting resins are irreversibly created during part production by building chemical bonds within the resin matrix and the fiber sizing. It is easy to see that understanding and controlling this curing process is crucial for the final performance of the composite part and you cannot make mistakes here.
Thermoset curing itself is a complex chemical reaction depending on the resin chemistry, as well as time and temperature as the reaction coordinates. It is therefore imperative to know the temperature influence on the resin cure. Due to thermodynamics, the temperature influence is exponential – i.e. a small change in process temperature, e.g. setting too low, will significantly decrease the performance of the part. The physical measure is the degree of cure of the matrix resin and this will later on influence all mechanical properties tested on the composite part.
The good news is that these relations are well known already, for example in Steve Sauerbrunn's article "Thermosets: How to avoid incomplete curing"2 and in every better polymer composite textbook: What it comes down to is that so-called TTT (time-temperature-transition) diagram of the material under use. It basically shows what will happen to your resin if you keep it at a certain temperature for a certain time. For example, if your curing temperature is too low – you will freeze the reaction due to vitrification before full cure. Furthermore, you can only reach full cure of your sample if the curing process temperature is higher than the highest possible glass transition temperature for that resin, otherwise you can wait for the part to cure forever. In the end if you understand the curing properties of your composite resin, only 5°C more could eventually speed up your reaction 20-30% - and this is when we start talking about time and money in carbon composite production.
There are of course many other material properties one has to know about and we can see the composite industry is measuring more instead of less. As a consequence of that trend, many people are implementing this knowledge into more and more sophisticated simulation programs that bring all different material properties together making it much easier to deal with the complex nature of the composite materials.
The last important challenge for composite manufacturing we would like to point out is the process of manufacturing itself. As stated previously the overall challenge is cost – so it would be the best solution if we get a manufacturing process to be fast, reproducible and efficient. Then we can start to scale up production.
Many high-volume production processes originate from other production techniques as for metals or plastic parts and needed to be modified to fit the properties of carbon fiber composites. The most prominent techniques are compression molding as used for the BMW i3, resin transfer molding (RTM) that is borrowing concepts from plastic injection molding as well as other molding techniques such as hand lay-up in a mold etc. Furthermore, there are slower, but more performance optimized technologies like AFP (automated fiber placement) and ATL (automated tape laying) available, that take the anisotropic nature fully into account as they place single tapes of fibers into the force direction where they are needed.
And this lack of knowledge is costing us time: we wait until we hope that everything is cured inside, aerospace industry adds hours of safety margin here, open it and take the part from the mold. In order to verify the curing process now we have to undertake an external quality control which uses up even more time and resources needed for measurements, verification and so on.
Now imagine that for a high-volume product line in automotive manufacturing– it is not going to work and is far too expensive.
How do we get out here? Imagine you would have a sensor telling you the exact state of cure throughout the curing process directly from inside the mold. Furthermore wouldn’t it be great to make the sensor talking to the manufacturing plant? Imagine you have overheated the oven but don’t have to worry about it: the following necessary steps to get a perfect composite part afterwards are going to be adapted automatically.
And this is when the NETZSCH dielectric analyzer DEA 288 comes into play: it has a sensor that can be related to the degree of cure.
The way the NETZSCH DEA sensors work is unique and we are currently advancing them to fit carbon fiber materials as well.
This is the overall problem that we can solve for automated process monitoring and control – even for a complex material such as fiber-reinforced composites. There are many research projects going in these directions already, such as OptoLight at AZL Aachen, at Fraunhofer ICT in Pfinztal and at aerospace and automotive OEMs as well. This will insure therefore the challenge of speeding up the composite manufacturing process itself is solved.
Nevertheless there is already good news:
That is what I wanted to show in this overview article. We are currently in an optimization phase of the composite market and that’s the time the tricky things are being discussed – and eventually solved. So the “hype” is over, let’s get back to work now. You should be confident that we’re on the right track.