Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Niobium-titanium rod radioactive?

  1. Nov 2, 2014 #1
    I just purchased a niobium-titanium rod, that I assume was once used in an accelerator experiment. Is there any danger that it might have a hazardous degree of radioactivity, from exposure to synchrotron radiation, assuming it was on the outside curve of a particle accelerator?

  2. jcsd
  3. Nov 2, 2014 #2


    User Avatar
    2017 Award

    Staff: Mentor

    Every piece of equipment that leaves a particle accelerator has to be checked and the radioactivity level has to be low enough for the material to leave the controlled areas. Where "low enough" means "you can carry it around all day without getting a significantly higher radiation dose".
    Unless something went really wrong, it won't be radioactive to any relevant degree.

    Particle physics does not look like the right forum... I guess nuclear engineering fits better even without a direct engineering question.
  4. Nov 2, 2014 #3

    Thank you for the response, that certainly allays my fears.
    Last edited: Nov 2, 2014
  5. Nov 3, 2014 #4

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    However, most materials from particle accelerators don't end up on Ebay. Without knowing the item's provenance, one cannot tell whether this was properly surveyed or not.
  6. Nov 3, 2014 #5
    My brother has a Geiger counter, so I'll check it with that. The low price caught my attention, as I had wanted to purchase niobium, or niobium alloy, in the past, but it would have cost several hundred dollars, as there was a minimum quantity requirement. I've been wanting to experiment with this Type II superconductor, and figured buying some would be a first step. The next step would be finding a facility, possibly a department within my old employer, that would allow me to bring the piece down to the desired temperature.
  7. Nov 3, 2014 #6


    User Avatar
    Science Advisor
    Gold Member

    Are you sure it came from an accelerator? It might just as easily have been part of the assembly a superconducting magnet; e.g. for MRI/NRM experiments.
    Also, NbTi has a Tc of 14K; you will need liquid helium to "play" with it.
  8. Nov 3, 2014 #7
    I actually don't know where it came from. If it's Tc is 14K, that's a little better than Wikipedia, which indicates 9.2K (http://en.wikipedia.org/wiki/Niobium-titanium). But, it's probably from some discarded device/experiment, and the individual pieces were likely sold surplus at a low cost, which the eBay user bought up a large quantity of.

    Just yesterday I came across a cooling process called "cryogen free dilution refrigeration". Several companies seem to make them: https://www.google.com/?gws_rd=ssl#q=non+cryogenic+dilution+refrigerator&spell=1 But it seems like there are only a few hundred of them in the world, so they are very specialized equipment. I didn't check prices, but assume it is astronomical, not something a hobbyist could afford.

    The 'play' part is simply to pass a high voltage discharge through a NbTi bar, maybe half an inch long, with a sensitive accelerometer nearby. I won't mention the purpose so as not to get in trouble with the moderators.

    Considering the difficulty of finding a place, with cryogenic facilities, to accommodate me on this experiment I may have to fall back on the YBCO superconductors, that are readily available to hobbyists. I actually have several on hand. These 1 inch diameter superconductors are always badly warped making it impossible to simply press electrodes to each side. But in a recent email from the owner of Superconductor.org I was told that he bonds silver epoxy to either side of his ceramic based superconductors for testing. He told me he cures them at 200 degrees for an hour, if I remember correctly. A while back I tried the silver epoxy approach, but it seemed to ruin the YBCO superconductor, as it never showed the Meissner effect again. I cured it outside in a humid marine environment, so maybe moisture penetrated the YBCO altering its chemical makeup. So, using his method might do the trick.
  9. Nov 3, 2014 #8


    User Avatar
    Science Advisor
    Gold Member

    Wikipedia is wrong in this instance, pure Nb has a Tc of 9.2K but the Tc of the popular alloys (NbN, NbTi) are actually a bit higher.
    A cryogen free refrigerator (a pulse cooler) that can get you down below that Tc of NbT would probably be £50 000 or so, perhaps a bit less. Single stage pulse coolers are significantly cheaper (a few thousand pounds) but can only go down to about 20K.
  10. Nov 3, 2014 #9
    I actually thought that since I previously read, some days earlier, that pure Nb was 9.2K, and it didn't seem right that an alloy would be exactly the same. Makes me realize that Wikipedia isn't always correct. Is there any metal that superconducts at the 20K threshold?

    Thanks for mentioning the prices on pulse coolers. I actually was wondering if some company, or institution, might have one in the Northeast United States, thinking they might be more amenable to an outsider running an experiment, since they wouldn't have the complications involved with using liquid helium.
  11. Nov 4, 2014 #10


    User Avatar
    Science Advisor
    Gold Member

    MgB2 (magnesium-diboride) has a Tc of 39K but is quite difficult to work with.
    Running experiments at cryogenic temperatures requires quite bit of knowledge about how to operate the equipment and handle the sample. So no, it it is not the kind of experiment they would let someone do without supervision.
    Also, cooling things down with helium is much, much simpler than using a cryocooler; at least if the sample is so small you can just attach it to the end of a rod that is then inserted into a dewar. Simple experiments can be done in 20 minutes, whereas with a cryocooler you would have to wait many hours to go down to 20K (and then wait several hours to warm it up). But again, not something you would let someone do without prior training.
  12. Nov 4, 2014 #11
    Thank you, f95toli, for clarifying the differences between using a cryocooler and a dewar of liquid helium. I was misled about the difficulties of using liquid helium from reading some other threads devoted to liquid helium, where someone mentioned that an open dewar of liquid helium would evaporate in one second, though they didn't specify how much, or how little, of the liquid helium would be involved in such quick evaporation.

    There are departments at my old employer which, I believe, utilize liquid helium. I'll ask a friend still working full time there to check on that.
  13. Nov 8, 2014 #12
    My NbTi rod arrived in the mail a few days ago. It's 6 and 1/2 inches long and 1/4 inch in diameter. It's not as heavy as steel, but the surface texture and color reminds me of steel. One end is cut, as with a hacksaw, or bandsaw, the other end looks like it was chopped with a metal shear. The whole thing is ever so slightly warped, the sheared end being mangled for half an inch. But if I cut a 1 inch section from the good side, it will hardly be noticeable. With my multi-meter it seems to have very low resistance, maybe a fraction of an ohm. I don't think this alloy is at all toxic, but I'll double check on the net. I assume it will cut with a regular hacksaw, though I'm aware that titanium is quite hard, as it's used in the Oceanographic industry where I worked for 30 years. My brother has a radiation detector good for all 3 types - gamma, beta, alpha, and I'll check it next weekend.

    Another thing I have to think about is whether I want to use regular Sn/Pb solder, or silver solder, to attach wires. I'm not sure either type of solder would adhere to this alloy. Also, I'm wondering if the solder bond would separate at liquid helium temps?
  14. Nov 20, 2014 #13
    I cut off a 1 inch section of my 1/4 inch niobium-titanium rod, but discovered it won't take ordinary 60/40 tin-lead solder. Since niobium-tin alloy is also a superconductor, and is partly composed of tin, presumably it would solder with tin-lead solder. But it's supposed to be very brittle, plus I have no idea where to buy a piece of it. If anyone has any ideas about soldering to niobium-titanium, or if niobium-tin would take ordinary solder, and where to find this alloy, please let me know.
  15. Nov 20, 2014 #14


    User Avatar
    Science Advisor
    Education Advisor

    This paper: http://www.osti.gov/scitech/servlets/purl/7215719 goes into fairly exhaustive detail about soldering a Cu clad Nb-Ti composite. The authors look to be concerned about wetting both the Cu and the Nb-Ti, and express concerns over how well the Nb-Ti was wet. Hopefully this gives you what you need.

    My only other idea would be to email someone who deals with Nb-Ti. Such as a company that makes superconducting magnets or an accelerator facility that uses them.

    Good luck!

    ETA: This http://www.assemblymag.com/articles/85409-soldering-the-unsolderable seems to suggest ultrasonic soldering.
  16. Nov 20, 2014 #15
    Thanks for checking into those strategies. Unfortunately my Nb-Ti bar doesn't have a copper cladding. Conceivably I might be able to have the piece clad in copper, then it shouldn't be difficult to solder, as all printed circuit boards are copper clad. I've also soldered to sheet copper, after careful cleaning. The ultrasonic seems like it would also work as both niobium and titanium are listed as substrates that can be soldered with this method.

    Another person privately emailed me and mentioned that TIG welding will also work. That might be the quickest and easiest route, so I'll check out some local shops that can do TIG welding.
  17. Nov 21, 2014 #16


    User Avatar
    Science Advisor
    Gold Member

    Nb can be TIG welded (with some difficulty), I am not sure about NbTi but I guess it might work,

    However, is there a reason why you can't use a mechanical connection? Nb, NbTi, NbN etc can be wire bonded using aluminium wire if you have access to a bonder. You can also make crimped connections (you can even make superconducting joints if you crimp two NbTi wires together ).
    If your piece is large enough you could even drill small holes and use screws. Soldering to Nb is practically impossible.
    Also, I still don't understand what you are trying to measure; if you only want to see if the piece is superconducting you could probably just glue some Cu wires to the surface using silver glue (use the stuff meant for repairing PCB traces if you can find any actual glue).
  18. Nov 21, 2014 #17
    Actually, there's no reason I can't make a mechanical connection, but I was hoping to avoid that complexity. Also, I was concerned that physical separation of the mechanical contact might occur as the whole structure was dipped into the liquid helium, and the parts shrink. If the contact was under pressure, via spring loading, the spring would have to be above the liquid helium, with beams going down into the dewar to transmit the mechanical force, as I assume a metal spring would lose its tension in liquid helium.

    Your mention of being able to bond aluminum to NbTi might be the ticket. A "bonder" just uses a very large current pulse to bond two pieces of metal, and my apparatus does just that - provide a very large current pulse. In the past I've accidentally bonded a steel screwdriver to aluminum rails. It actually made a pretty good bond, requiring a lot of mechanical force to separate it. But that makes me wonder is there something special about aluminum that it can be bonded to NbTi, and other metals can't be bonded to NbTi? I see that aluminum is in a different part of the periodic table than Nb or Ti, the latter two being in the same section of the periodic table.

    I read about the toxicity of niobium, and it sounded scary - liver and kidney damage if ingested, or breathed in. So I wanted to do the minimum of mechanical processing, to avoid contamination of my workplace with the shavings and dust. I thought about drilling and tapping the ends of the rod, something I've done many times before on other projects with more convention metals - steel, aluminum, but it would contaminate my drill press area.
    Of course if I used a high voltage discharge for bonding to the NbTi it would probably produce a small amount of niobium vapor. But I could do that outdoors, as my system is battery powered, with a 12 volt lantern cell.

    Funny, I didn't even think of using silver epoxy to bond wires to the ends of the NbTi rod. I used silver epoxy once on a YBCO, 1 inch disc, but it ruined the superconducting characteristics of the disc. However, I let it cure outdoors on a dock, near my workroom, in a humid marine environment, so moisture may have penetrated the YBCO, altering its chemical properties. I was always careful to keep my YBCO discs in a plastic container with lots of desiccant bags to soak up any moisture. But, recently, the owner of Superconductor.org, told me that he uses silver epoxy all the time to bond wires to the ceramic superconductors. He told me that he cures his samples in an oven at 200 degrees for 20 minutes. So, I may abandon efforts to work with Niobium alloys and just go back to YBCO discs, which are readily available to hobbyists.

    As to what I am trying to measure, that gets into pseudoscience. A materials scientist, named Evgeny Podkletnov, claimed that he detected brief acceleration pulses (1/10,000th of a second), as large as 1000 g's, from a YBCO superconductor subjected to 2 million volt discharges. Another group, led by Martin Tajmar, at the Austrian Research Center (ARC) reported small (100 micro-g) acceleration signals when rapidly spinning up a small niobium ring inside a liquid helium cryostat. The outer part of the ring was subjected to only 7.68 g's maximum. So, I thought maybe acceleration of the condensate inside these superconductors is what leads to acceleration signals. I also wondered since niobium has 10 times the cooper-pair density as YBCO whether it might enhance such 'signals', perhaps in a direct proportional manner. Hence the desire to use niobium or its alloys. NASA checked into this a long time ago. But, when I read their report, it seemed like they just used a perfectly stationary YBCO disc, about a foot in diameter, and looked for a diminuation of gravity above it with possibly the world's most sensitive gravitometer. They found nothing. The ARC group later retracted their claim. So, admittedly, this is all on rather shaky ground. But, I thought, if the degree of acceleration of the condensate was perhaps directly proportional to the 'signal', then I could greatly improve on the ARC teams magnitude of acceleration by the high voltage method. I'm assuming the cooper pairs would undergo a brief acceleration spurt much larger than 7.68 g's when subjected to 600 volts.
  19. Nov 21, 2014 #18


    User Avatar
    Science Advisor
    Gold Member

    No, a bonder uses a tungsten tip that vibrates at ultrasonic frequencies to "dry weld" a wire to a surface, there is no current involved. It is standard method in e.g. the semiconductor industry, most labs have semi-automatic machines for bonding samples.

    YBCO is very sensitive to moisture and the silver glue was most certainly not to blame; it is used all the time for connecting wires to YBCO and other materials

    Nb is pretty harmless unless you eat it; remember that is a very popular metal for making jewelry meaning touching it is not dangerous at all, .
  20. Nov 21, 2014 #19
    Thanks for the information on the lack of toxicity of Nb, as long as you don't eat it. I didn't check titanium, but working in the oceanographic industry I came across it a lot and handled it many times. I think I even tried cutting a piece once. I don't have a lathe, plus the 1/4 inch piece is a bit small for drilling into, so the bonding approach sounds good. I wonder if bonders are inexpensive enough for hobbyists to use at home? If not, I live in New England, and maybe there's a lab I could contact in the Boston, or Worcester area.

    I watched a couple of videos this morning on copper plating. One fellow used a hot solution of copper sulfate. But he had previously nickel plated the coin that he copper plated: Here's the video:

    I'm not sure the same procedure could be used on NbTi. But if it can, that would solve everything, as copper is readily solderable with ordinary Sn/Pb 60/40 solder.
  21. Dec 3, 2014 #20
    Being that I didn't have a lathe, I had a local machine shop drill and tap for 6-32 thread size, the ends of a 2 inch piece of my 1/4 inch diameter niobium-titanium bar. Attaching electrodes will now be a cinch.

    But I have a question, more academic than practical. Is it possible that small numbers of cooper pairs might spontaneously arise in niobium, or its alloys, at liquid nitrogen temperatures, even though LN2 is at 70 Kelvin versus the actual superconductivity temp. for niobium at 9.3 Kelvin? I realize that a handful of cooper-pairs among gazillions of lattice atoms would not lead to detectable superconductivity, but wondered if there was some finite probability of cooper-pairs arising at temperatures above niobium, or its alloys, critical temperatures.

    Also, I was wondering if niobium can be alloyed with copper? The reason I ask is that all, or most, of the ceramic high temperature superconductors are cuprates, so it seems copper atoms play a role in the superconductivity seen in these ceramic compounds. Looking at diagrams of the lattice structure of these ceramic superconductors, it seems that they are layered, such that the copper atoms appear to be arrayed in a plane. That mechanical arrangement might be crucial for their high temp. superconductivity. This structuring might not be the case for a bi-metallic compound of niobium and copper, assuming such an alloy is even possible. So, maybe, there wouldn't be any improvement in the superconducting temperature.
  22. Dec 3, 2014 #21


    User Avatar
    Science Advisor
    Gold Member

    No, the critical temperature is when the Cooper pairs first start to form; you actually have to go well below Tc in order to reach a temperature where nearly all the electrons are paired up (about Tc/10 and below). This is why Nb isn't very useful until you are at a temperature of say 4.2K; at higher temperatures you have so many normal electrons around that the material is e.g. still quite lossy.

    The layering is certainly crucial; but we don't understand exactly what is going on. There is no simple correlation between how an element behaves on its own and how it would work as part of a more complex structure such as the copper oxides. Hence, it may or may not be possible to incorporate Nb in a high-Tc superconductor, but whether or not is would give a reasonable Tc would have nothing to do with the fact that the Tc of Nb on its own is high, the mechanisms involved are very different.
  23. Apr 27, 2016 #22
    I just ran the experiment, mentioned up-thread, where a high voltage pulse (just over 500V) was passed through the length of a 2 inch long, by 1/4 inch diameter, Niobium-titanium (Nb-Ti) rod, immersed in liquid nitrogen. In direct alignment with the rod was a one milli-g resolution, micro-chip (ADXL203) accelerometer, enclosed in an aluminum Budbox, (for RF shielding). The discharge cycle was repeated 10 times, with the position of the micro-chip containing Budbox being slightly adjusted with a threaded rod driven X-Y positioning system that I recently built. The purpose of these small adjustments was to have different parts of the Nb-Ti rod's circular cross-section come into alignment with the 'target' accelerometer-microchip, which is very tiny.

    The motivation for conducting the experiment were claims by E. Podkletnov and C. Poher of detection of brief acceleration pulses, (collimated with the current axis), when high voltages were discharged into, or through, YBCO superconductors at liquid nitrogen, or liquid helium temperatures. It was their opinion that the current pulse interacts with the condensates within their superconductors to produce these claimed signals.

    I did not have a temperature low enough to form a condensate within my niobium-titanium rod, so I had no expectation of detecting any acceleration pulse, and in fact nothing was detected in any of the 10 runs. The critical temperature of Nb-Ti is 14K, while liquid nitrogen is around 77K. My electronics is set up such that the scope is triggered for a single sweep at the moment of the high-voltage discharge. The 'target' accelerometer-microchip is located 3.625 inches from the near end of the Nb-Ti rod. Calibration tests using a spark-gap at the same distance, clearly showed the arrival of the sound impulse, with a crisp vertical spike at 280 microseconds, followed by a longer trailing edge. This is within about 1% of the velocity of sound at ambient conditions, so I know that my system is working just fine, and can easily separate any acoustic impulse from a signal, that if it actually exists, would be expected to propagate at c.

    The final experiment will require lowering the Nb-Ti to liquid helium temperatures to allow formation of a condensate, and then subject it to high voltage discharges. To that end I plan to contact Universities in the Boston area, to see if they would accommodate me on running such an experiment.
  24. Apr 27, 2016 #23
    I would like to post pictures of my experiment without linking to my website. The places where you can upload photos on the net I find difficult to use, and slow, because of ads, so I'm wondering if it would be OK to create a page on my website, that has no links to the rest of the site? That would be extremely easy for me to do.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted