I am working with a set of alloys that must first be ground up in a ball mill at cryogenic temperatures. It is important that no contamination makes it into the material and careful steps are taken to ensure this end.

Unfortunately when we ran a set of samples through our RBS lab and then again through PIXE we saw that we had Fe and Cr contamination. At first it was thought that it came from the cutting tools used to prepare the samples so we ran another sample that was cut using EDM. The results were the same. The only other contact the material has with a stainless steel is from the ball milling step.

We have been using a 440C for the cup and balls but it would seem a 316L may be a better fit for this application. I know 440C does not typically do well at cryo-temps but that is what other labs have been using so we didn't expect for there to be problems of this type.

Some things to consider for the new cup and ball material are machinability, cost, availability, cryogenic properties, vibration and fatigue resistance, and ability to be sealed (cups are filled in an inert Ar atmosphere). Another possibility is a heat treatment of the current 440C materials, however I am unsure what the best approach would be in that regard.

  • 1
    $\begingroup$ Implausible sounding suggestion that may be useful (if it works I'd be pleased if you remember who suggested it :-) ).: See if UHMWPE can do anything for you. While it hardly sounds like a ball-milling ball solution in its own right, it may have a role in the cup, andmaybe in the ball makeup. It's as 'tough' as all getout and retains its properties at deeper cryogenic temperatures better than about anything else otherwise similar. It's good in chopping boards, body armour, tugboat buffers, dragline buckets, indecently cold temperatures, and, maybe also, cryogenic ball mills. $\endgroup$ Commented May 23, 2015 at 9:34
  • 1
    $\begingroup$ I am reading about your material suggestion; UHMWPE has some excellent mechanical properties. I have worked with a few polymers and in fact the cryo-chamber which houses the cup assembly we made out of nylon. It seems UHMWPE has many applications although begins to experience brittle failure at extreme temperatures below -150C; we are running around -195C(ish). Your material suggestion is inexpensive and readily available, I may have to pick some up to play with :) $\endgroup$ Commented May 23, 2015 at 13:55
  • $\begingroup$ Given that they are already generally brittle, even at room temperature, have you considered ceramic materials for the balls? Silicon nitride is a potential candidate as it is extremely hard and has seen extensive use in bearings as a low-wear material. The best part is that because they are already widely used in bearings, silicon nitride balls are readily purchasable as-is at many sizes. $\endgroup$ Commented Oct 1, 2015 at 17:33

3 Answers 3


If cost isn't an extreme object, then hou might have lowest contamination by using a Co+WC makeup for your balls, and use a cryoplastic cup to avoic abraiding between balls & cup.

The WC is extremely thermally stable & has an incredibly high surface hardness (not to mention density). As long as your Co binder could hold up to the stresses (or you could find another suitable binder), you should enjoy much lower contamination, along with accelerated milling times/efficiency due to the increased density & hardness of your milling balls.

WC is readily available in powder form, ready to be formed & sintered using any machinery capable of heating & injecting molten Co (or Ni) to 'wet' it. Ofc, if the Co proves untenable due to contamination or processing costs, you could always use a low-outgas, cryo-rated epoxy to wet/bind the WC powder (might work better anywise, as I think about it more).

  • $\begingroup$ I agree that WC (tungsten-carbide) could be an excellent choice. To expand on your concerns about how to address material transfer, I've had carbide tools PVD coated, which could be good option in this case. The nice thing about PVD coatings is they are usually a different color from the base metal, which means it's easy to see and track wear which can be used to establish how long balls last and put in place a PM routine to change them out before they wear to the point of material transfer. Most coatings can be removed and reapplied as well - keeping maintenance costs down :) $\endgroup$
    – CBRF23
    Commented Apr 27, 2016 at 19:45
  • $\begingroup$ An interesting idea, although it would take some work to build this setup. We use nylon for our outer shell that houses the cup assembly to flow our liquid nitrogen through. It is easy to machine; at least in comparison to 440C (although we are getting good at it!). As far as the WC powders, I have not come across any equipment on campus that could perform this operation. We built a hot isostatic press (HiP) that is used to compress powders under high temperature, however we designed it not to exceed 500C (and even then only for short period of time to avoid issues with creep). $\endgroup$ Commented May 24, 2016 at 14:28

I debated between Answer or Comment on this, but ultimately I think it is more of an answer - albeit an imperfect answer.

It sounds to me like the primary problem you are trying to solve is preventing material transfer from the balls/cups etc. during the ball milling process. I don't think you necessarily need to change materials, and perhaps could even get away with just adding a PVD or CVD coating to the existing components.

DLC (diamond-like-carbon) coating pops to mind first,

however I think there are many coatings that could potentially serve you well. DLC coatings are very hard and wear resistant (as the "diamond-like" name would suggest). They do not flake or chip off, and while I don't know the end use or the nature of the "no contamination" requirement, these coatings are completely inert to almost all organic and synthetic chemicals, and completely inert to the human body as well.

For this application, I would think a ta-C or possibly a ta-C:H coating could work well. Another DLC that is supposed to be extremely hard and wear resistant, but that I have no personal experience with, is UNCD (ultrananocrystalline diamond).

You may also find more traditional tool coatings, such as TiN or TiAlN could work well for you - I couldn't make a recommendation as to the best coating for your application. I'm not a coatings engineer, just an engineer that has had good experience using these types of coatings for my own applications in the past :)

The main benefit of 440C is it's high hardenability

With a maximum hardness of around 60 rockwell C, 440C is a stainless that can rival many tool steels.

It sounds like you are using this material in the annealed condition, which is questionable to me. This material is typically chosen for it's high hardenability - it's not (to the best of my knowledge) commonly used in the annealed state.

It sounds like you chose this material because it's commonly used in other similar designs; I wonder what condition is this material used in those designs.

If you can get your hands on some 440C components from one of those other designs I would do a rockwell test to see whether or not it's heat treated. I would put money on it being heat treated. Annealed material will probably be in the 20's-30's rockwell C, while heat treated material is going to be in the 50's.

If you do intent do to a rockwell test, try doing in on a component with a flat as spheres can be difficult to test and give inaccurate readings.

For a place to start with heat treat, I find Carpenter's data sheets are usually pretty reliable. Their recommendations for 440C are:

  • HARDEN: Heat to 1850/1950ºF (1010/1066ºC); soak; quench in warm oil or cool in air. Hardness will be ≈60HRC. Do not overheat or you will not be able to attain maximum hardness.
  • TEMPER: To remove peak stresses and yet retain maximum hardness, temper at least one hour at 300/350ºF (149/177ºC).

If you are looking to heat treat, I would probably start there.

I do not think 316 would be a good choice

As it's a much softer (gummier) material than 440C, and I think it would exasperate the material transfer problem.

Now I said this was an imperfect answer,

because I don't know that it directly answers your question. It offers an avenue to a possible solution, however it is incomplete as you'll need to discuss with a coating expert the exact needs of your application and see what they say.

For example, I have no knowledge of how these coatings perform at cryogenic temperatures, or what coating would work best with the abrasives used in your ball milling process.

I know I have used DLC coatings to solve some unique problems where I work, and they have allowed me to do things I don't think I could have found an alternative for.

I also wanted to say that finding a reputable and reliable coating vendor was the hardest part of adapting this technology for me. As that can be quite a daunting challenge, I offer a suggestion only as a place to start, and with no further endorsement other than I've personally had good experience with Oerlikon Balzers. I do not know the policy on recommending vendors, and I have no affiliation with any vendors - so feel free to edit this answer to remove the vendor name if it violates any policies.

Regardless of who you choose to do business with, I would highly recommend speaking with one of their applications engineers about the specific needs for your application, and seeing what coating they recommend.

Closing thoughts

PVD/CVD coatings are tribological in nature - they affect the interface with other materials, but do not change the properties of the base substrate.

When we talk about bearing design, there is a property commonly referred to as embedability. This basically refers to a bearing materials ability to absorb (or embed within itself) foreign materials.

Without knowing more about the nature of your design, my gut tells me that you would not want this to happen, as it would mean your balls/cups etc. would be embedding some amount of the alloys that you were trying to grind. It seems unlikely to me that this would be desirable. This is another reason I think 316 would be a poor choice.

To reduce embedability, you want a harder substrate.

My recommendation to you would be to heat treat the 440C components first, then apply a PVD coating afterwards, to obtain the best combination of properties for your application. I would definitely talk to a coatings engineer about what coatings will work for your application (e.g. temperatures, material compatibility, etc.)

  • $\begingroup$ There are some excellent suggestions here. The latest idea we had (besides regular replacement of the 440C balls) is to switch to a nitronic 60 for the balls and maintain the 440C for the cup assembly. The N60/440C combination has very low incidence of microspallation at cryogenic temperatures. The coatings are excellent suggestions, I will look into this further. If we decide to go this route, and are successful, then I will come back to vote your solution as the correct answer. This may take some time... :) $\endgroup$ Commented May 24, 2016 at 14:50

I recommend alumina balls (not steel alloys though).

I used ball mills in 5 ton capacity and pot mill in lab (250 grams) using alumina balls.

I think alumina balls are inert (already oxidized) so that even though they may introduce some impurity, I don't believe this will affect your alloy chemically like Fe and Cr will.

With alumina balls, the rate of wear is very low. Everyday use requires only a regular top-up of balls every 3 months.

Alumina balls also can be used for cryogenic temperature (CMIIW). Something to keep in mind though: pot mill temperature will be very hot (almost boiling) after milling, so it may prove difficult to maintain cryogenic.

  • $\begingroup$ I will try this, if it works then you will get the correct answer. I am not sure what our internal temperature is, although we know from electron microscopy that the materials cold weld themselves during the milling process. The desired grain size is around 50nm for the materials we are currently working on, and the current setup does this well (minus the contamination). We have kicked around the idea of installing a thermocouple; it would be a terrific engineering challenge. $\endgroup$ Commented May 24, 2016 at 14:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.