Experimenting with Vacuum Resin Infusion and Basalt Fiber

During this time of pandemic and business closure, when I’ve been unable meet with clients as usual, and trapped inside due to fire and smoke, I have explored some techniques for using new materials to make composite shoe parts. I’m working with basalt fiber and epoxy. Below are my initial trials and some details on how I made them.

Basalt fiber test parts

Background and Back Burner

I first tried to make carbon fiber parts when making prototypes for a pair of bicycle touring boots requested by a client. 10 years ago when I visited Ortho Baltic (a Lithuanian company where I buy some of my lasts), I was impressed with their ability to produce a wide range of very high quality carbon fiber parts. I contracted Ortho Baltic to make the carbon fiber midsoles for me.

Prototype of carbon fiber midsoles with slots for SPD cleats.

Ortho Baltic can make such high quality parts because they are cured under vacuum in carefully controlled high temperature ovens. Because making custom shoes means never doing the exact same thing twice, I wanted to see if I could develop the facility to make custom carbon parts on par with Ortho Baltic when needed, but with less equipment. The fact remains that carefully controlled heating and cooling is really the only way to produce these kinds of parts.

Without using ovens, I’ve found that the next best thing is to use vacuum resin infusion. I watched videos on this over the years but kept the idea on the back burner because I was never excited about working with epoxy and carbon fiber. Epoxy is considered non-toxic, unless you sand it and inhale the dust. Carbon fiber is very dangerous to work with. When cut or sanded, particles of carbon fiber float into the air and get everywhere.  Fibers can become embedded in your skin, eyes and lungs. Carbon fiber’s tiny particle size  makes it an inhalation hazard.  After carbon fiber has been set with resin, it can make splinters (like fiberglass) which make the parts no fun to handle. The coatings used on the fibers are often toxic as well (OHS Safety Sheet). Yikes. It’s enough already to do all the things needed to make custom shoes without also becoming an expert on plastics and producing custom composite parts. Nonetheless, my curiosity remained.

Front Burner

Vacuum resin infused prosthetic leg socket.
Vacuum resin infused prosthetic leg socket.

In the summer of 2019 I visited one of my students at his shop in the Shriner’s Children’s Hospital Orthopedics and Prosthetics lab. There I saw vacuum resin infusion used to fabricate all kinds of prosthetic devices, and that’s where I learned about basalt fiber. Basalt fiber is way less dangerous to work with than carbon fiber.

Braided basalt fiber sleeve set on insole before infusion.

In August I finally started gathering the basic materials for a vacuum infusion setup. My first goal was to make a midsole which would provide some relief to a stiff big toe and ease toe-off. It is intended to be a combination of rocker bar, spring, and shank – all in one piece. Using the insole with a 1cm inset as a template, I covered the entire area using one basalt fiber braided sleeve. Since it’s a tube, the sleeve is equivalent to two layers of the same type of fabric.

The braided sleeve format means you only need to cut it to length – no fraying edges. It can be shaped easily without coming apart. If you could get a vacuum seal with the upper, the part could be made directly on the last. I was unable get a vacuum seal with this upper. There are a staggering number of places an air leak might occur, but I still think a vacuum seal with the upper on a last is possible.

Maximizing Green Stage

Since getting a good vacuum seal is so difficult, I decided try to keep my setup as simple as possible. I made a flat mold and transferred the part to the last before it has completely cured. If you get it at the right time, the epoxy will be cured enough to keep the part together as a unit, but also sticky enough to make a decent bond to the insole. This is also known as “green stage.” In green stage you can cut the part with scissors or a knife, if needed, which also saves you from sanding a too-big part. This also means I didn’t have to make a matching pair (left and right) of the much more complicated compound curved mold of the last bottom.

Using rubber cork, I made a mold for the part on a rigid piece of ABS plastic (4mm thick was the minimum). The base has to be rigid enough or it will bend under vacuum if the bagging film pulls on the perimeter.

With the film applied and tubing to pull the vacuum and let the resin in, it only takes a minute to actually infuse the fabric with epoxy and make the part.

Vacuum pot used as a resin trap, the "bagged" part, and a silicone cup with the epoxy on the inlet side.
Vacuum pot used as a resin trap, the “bagged” part, and a silicone cup with the epoxy on the inlet side.

So little resin is used to infuse the part that the vacuum changes hardly at all. The resin trap provides enough of a stored vacuum that you don’t need to have the vacuum pump connected to complete the infusion. It takes hardly 2 minutes to infuse. On the second run, I infused it much more slowly. Here’s about 13 minutes sped up to about a minute and a half.

Infusion – The Movie

Sped up version of infusing the part slowly.
3 hours after I first mixed the epoxy, here’s the transfer to the shoe.

Recap and Plastic Waste

To revisit the pieces pictured at the top of this article, different vacuum levels, or none at all, produce very different results. Looking at the four pieces above, and starting from the left:

    1. Using a two part mold to squeeze out the extra resin made a very nice looking part that felt good to handle and was quite strong. The extra resin provides a lot of strength, but adds extra thickness and weight.
    2. This part had a leaky vacuum seal, which is similar to using very low vacuum. More resin is allowed in and the whole assemblage is allowed more space for both air and resin. The resin saturation and thickness is not very consistent, but it is strong and functional
    3. The product notes for basalt fiber will tell you that 30in/Hg vacuum will result in a part that’s a bit too dry, or “starved.” 20 – 24in/Hg is recommended. I wanted to see what “starved” looks like. The part is very thin and light, but brittle. There is not enough resin for the part to maintain integrity and it can easily be broken.
    4. 20in/Hg worked well for me, and I would consider even 15in/Hg. In making this last part, I tried  a two-step infusion – first infusing under full vacuum (30in/Hg), then releasing more resin with the vacuum set to 20in/Hg. The fabric is fully saturated, strong, light, and not breakable.

Sometimes you want a stiff part, sometimes you want something flexible. Sometimes you want a bit of both. Using a braided sleeve, the irregularity of the insole shape naturally creates stiffness in the shank and flexibility in the forefoot. It does this simply by narrowing and condensing the material in the shank area and spreading it out in the forefoot. More fabric can be added as needed to make the part stiffer overall or stiffer in places.

I’m already able to make parts without the need for flanges or other fabrication artifacts to be cut, sanded or otherwise discarded. My biggest challenge right now is trying to figure out how to develop ways to reuse the molds and setup materials.

When working with plastics and composites, I’m always shocked at the prevalence of single-use materials. For example, the plastic film that covers the mold is thrown away after one use. It’s difficult to remove it without destroying it, even though it’s quite tough. Bleeder cloth and infusion mesh are also not easy to re-use. A failed part is a genuine failure – there’s literally nothing other than the experience to be recovered and reused.