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

How does one acquire isotopically pure elements?

  1. Dec 27, 2006 #1
    How would one acquire an isotopically pure sample of an element? Are some available for direct purchase, or would it have to be done in-house with some sort of large centrifuge apparatus? (Im thinking about transition metals, not actinides)
  2. jcsd
  3. Dec 27, 2006 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    One can obtain 'chemically pure' elements, but 'isotopically pure' or enriched is much more difficult. One could use ultrahighspeed centrifuge technology, mass spectrometry, or atomic-vapor laser isotopic separation or a more recent version of LIS, known as SILEX ( http://en.wikipedia.org/wiki/Silex_Process ).


    Li, B, Zn and U are elements 'enriched' to increase the proportion of a particular isotope well beyond the natural proportion, but they still have more than one isotope.
    Last edited: Dec 28, 2006
  4. Dec 27, 2006 #3


    User Avatar
    Science Advisor
    Dearly Missed


    Interesting - do you know how SILEX differs from the LLNL-developed AVLIS?


    AVLIS uses a monochromatic laser beam tuned so that it ionize U-235 but
    not ionize U-238. The vapor of selectively ionized U-235 / neutral U-238 is
    then passed through an electric field to divert the U-235 and U-238 into
    separate paths. From the above link:

    "The absorption lines of 235U and 238U differ slightly due to hyperfine structure; for
    example, the 238U absorption peak shifts from 5027.4 angstroms to 5027.3 Å (502.74
    nanometers to 502.73 nm) in 235U. AVLIS uses tunable dye lasers, which can be precisely
    tuned, so only 235U absorbs the photons and selectively undergoes excitation and then
    photoionization. The ions are then electrostatically deflected to a collector, while the
    neutral unwanted uranium-238 passes through."

    Dr. Gregory Greenman
    Last edited: Dec 27, 2006
  5. Dec 27, 2006 #4
    Hrm, thanks. I guess I should be more clear, I suppose enriched (>95%) would be a much better way to put it.

    The reason Im asking is because I have been reading about medical isotope production. Some of the precursor elements have multiple isotopes, and literature indicates that a highly enriched target was used but they never go into detail on how it was created. Do you suppose most of the metals are just enriched at the production site? Or can you buy them elsewhere?
  6. Dec 27, 2006 #5


    User Avatar
    Staff Emeritus
    Science Advisor

    Yeah - I've been wondering about that myself. Ostensibly they are similar, but apparently SILEX works more efficiently than AVLIS, but I don't know how or why.

    GE has licensed SILEX and will build a pilot demonstration plant.

    Here is more - http://www.silex.com.au/s03_about_silex/s30_1_content.html [Broken]

    I'd like to know the details.

    Apparently they enrich Si, C, Gd, as well as others.

    http://www.silex.com.au/s11_technical/content.html [Broken]
    Last edited by a moderator: May 2, 2017
  7. Dec 27, 2006 #6


    User Avatar
    Science Advisor
    Dearly Missed


    I don't know how they could make that claim, because I do know that
    the efficiency of the AVLIS process is classified.

    The AVLIS copper-vapor laser pumped dye lasers can also be tuned to
    a multitude of frequencies. That's how the system was adapted to
    excite sodium in the upper atmosphere for the "laser guide star":


    The above link also shows a picture of an AVLIS-based facility for the
    production of medical isotopes; which is what our original poster was
    interested in.

    Dr. Gregory Greenman
  8. Dec 28, 2006 #7


    User Avatar
    Staff Emeritus
    Science Advisor

    Apparently there is a bilateral agreement between US (DOE) and Australia (ANSTO) to share nuclear-related technology. USEC and Silex Systems Ltd (or their predecessors) apparently established some relationship back in 1996, so I expect that they had some cooperation or technical exchange.


    USEC abandoned AVLIS because it was presumably not cost effective (?). Instead they focused on advanced centrifuge technology - and ATK is making the centrifuge tubes.
    Last edited: Dec 28, 2006
  9. Dec 28, 2006 #8


    User Avatar
    Science Advisor
    Dearly Missed


    Even WITH an agreement, as a matter of U.S. LAW; classified information
    can NOT be shared with a foreign commercial entity.

    Here's the text of the USEC press release when they discontinued AVLIS:


    Dr. Gregory Greenman
    Last edited by a moderator: Apr 22, 2017
  10. Dec 28, 2006 #9
    Since most medical isotopes are radioactive (some such as contrast dyes aren't), and these decay, it's impossible to get an isotopically pure radionuclide. You can get some that are >95%, but these would need be long lived.
  11. Dec 28, 2006 #10


    User Avatar
    Science Advisor
    Dearly Missed


    The isotope that you want to be pure isn't the one that's put into the
    patient. The isotope that you would like to be pure is the one that's
    put into the REACTOR!!!

    For example, you would like pure Molybdenum-98 to put into the reactor.
    Upon irradiation, some of the Mo-98 will be turned into Technicium-99m.
    Upon removal from the reactor, the sample will be a mixture of Mo-98 and

    One can then chemically separate the Tc-99m from the Mo-98; and the
    Tc-99m is the short lived isotope that is then put into the patient.

    If the Molybdenum that was put into the reactor was not pure Mo-98;
    then you get additional unwanted nuclides from the transmutation of
    the other Mo isotopes.

    Dr. Gregory Greenman
  12. Dec 28, 2006 #11
    Yeah, what Morbius said. I was looking at Lu-177. If you use natural Lu, you get unwanted isotopes. If you use enriched Lu-176, you still get a contaminant, the Lu-177m. If you get close to pure Yb-176, you can irradiate, wait for the Yb-177 to beta decay to Lu-177, and then seperate the Lu from the Yb via chemistry
  13. Dec 28, 2006 #12
    Don't you end up with some other isotopes of Tc? And why is the efficiency of ALVIS classified?
  14. Dec 28, 2006 #13


    User Avatar
    Science Advisor
    Dearly Missed


    If something is classified; the reason is usually classified too.

    Otherwise, that would tell you something about the thing that is

    Enrichment technology is very sensitive.

    Dr. Gregory Greenman
    Last edited: Dec 28, 2006
  15. Dec 29, 2006 #14
    Yes, that's true. But most medical isotopes are fission products rather than activation products.
    I assume you mean the reaction is 98Mo(n,gamma)99Mo which then immediately starts decaying to Tc-99m (about 87% branch ratio) and Tc-99 (13%). Then you get a mixture of Mo-98, Mo-99, Tc-99, and Tc-99m once the Mo-99 starts decaying. My point is that if you want an isotpically pure radionuclide, it's impossible. Once it becomes radioactive, it starts to decay to something else, and you now have a mixture of the parent nuclide and daughter nuclide.
    However, Tc-99m in the medical industry is exclusively made by 4 manufacturers. These manufacturers make Mo-99 generators by separating out the Mo-99 that is a fission product from other fission products. It's not 100% efficient, so there are some impurities. The neutron absorption cross section for Mo-98 is only 127 millibarns, so it's not very efficient to use neutron activation as a means of production. Regardless, it's still impossible to make it isotopically pure Mo-99 because once that first decay event happens, there is some Tc-99 (or Tc-99m) in the mix.

    Again, since it is so short lived, it's impractical to separate out the Tc from the Moly unless the hospital happens to be right next to the reactor.
  16. Dec 29, 2006 #15


    User Avatar
    Science Advisor
    Dearly Missed



    Medical isotopes are NOT fission products. Medical isotopes are the capture
    products of irradiation by reactors of a precursor. If you used fission products
    as the seed material; you would have a whole host of radioisotopes that you
    didn't want. Many of these would be isotopes of the same element as the one
    you sought. The only way to separate the isotopes would be an isotopic
    separation process.

    However, you would be doing this on radiologically "hot" material. That's complex,
    expensive, and unnecessary. You do the isotopic separation BEFORE you irradiate.
    You isotopically separate out the stable isotope, that when irradiated, will give you
    the desired target radioisotope. [If the desired precursor is a large fraction of the
    natural abundance; then this isotopic separation step is not neccessary.]

    If one were to scavenge medical isotopes from fission products; i.e. spent fuel -
    then one would need to do so shortly after the fuel was removed from the reactor.
    Spent fuel fresh from a reactor is too radioactive to handle without the massive
    facilities for reprocessing as were to be found at Hanford and Savannah River.
    Those facilities weren't called "Canyons" for nothing - they are truly massive.

    There's simply no need to scavenge medical isotopes from the "whitch's brew"
    that are fission products. Not when one can make medical isotopes directly
    by irradiation in research reactors or the High-Flux Isotope Reactor [HFIR] at
    Oak Ridge:

    http://web.ornl.gov/sci/rrd/pages/wedo.html [Broken]

    Why would you "want" something isotopically pure to put into the patient?
    You have this misunderstanding that we want something isotopically pure
    for the patient. No - you want it isotopically pure for the reactor.

    What goes into the patient doesn't need to be isotopically pure. If you had
    a mix of of Tc-99 and Tc-99m in the patient, that's no problem!! As long
    as the dose from the fraction of Tc-99m is enough. The fact that some
    stable Tc-99 is "tagging along" is not a big issue.


    Those manufacturers prepare Mo-99 "cows" from Mo-98 targets that have
    been irradiated in reactors. For example, when I was a graduate student at
    MIT, one of the jobs that the MIT research reactor does is to do these

    If one were to separate radionuclides from fission products, one would have
    to chemically reprocess spent fuel. However, reprocessing spent fuel in the
    USA is FORBIDDEN by LAW!! The reason is that another of the byproducts
    of fission reactions in the fuel is Plutonium. Plutonium that is not co-mingled
    with other radiologically "hot" isotopes is a nuclear weapons proliferation risk.
    That's the reason for banning reprocessing on spent fuel.

    When you irradiate a sample of Molybdenum in a research reactor like the
    MITR-II, you obtain a sample which consists of the desired radionuclides
    and some of the original material. You don't get fission byproducts like
    Plutonium; so there's no proliferation risk.

    Perhaps you thought the feed material for the 4 companies was spent fuel.
    However, that's incorrect. Their feed material are targets of a precursor
    element that has been specifically irradiated in a reactor with neutron
    irradiation facilities - like a research reactor.

    First NOTHING is 100% efficient, and that's fine because it doesn't have to
    be. As far as the cross-section for Mo-98 being a little more than a tenth of
    a barn - SO WHAT!!! The neutron fluxes available in a research reactor are
    quite high, and one can leave the target in the reactor for a long time and it
    doesn't affect the reactor to any appreciable degree.

    Regardless of how "efficient" you think it is; that's how it is done!!!

    You are a veritable FOUNT of MISINFORMATION today!!!

    What the hospital has is a "Tc-99m generator". It has at its heart a bunch
    of Mo-99 that is constantly decaying into Tc-99m. That Tc-99m is also
    decaying away - however at any given instant, there is a certain amount of
    Tc-99m in the generator. It is THAT Tc-99m that the hospital taps off when
    they need Tc-99m to give to a patient:


    "Tc-99m is a versatile scanning agent that is often considered the workhorse of nuclear
    medicine. It is obtained by elution from a generator ("cow") that contains the radioactive
    parent of Tc-99m, molybdenum 99.

    The generator is simply a column containing a resin to which Mo-99 is attached. The Mo-99
    decays to produce the short-lived Tc-99m (6 hr half-life). To obtain the Tc-99m, a solution
    (the eluent) is injected into the top of the column - the shield plug for the top of the column
    can be seen in the two photos to the right. The Tc-99m comes out the bottom of the column
    into the sterile collecting vial seen in the above photo. The collecting vial has a short
    breather needle to allow air out of the vial as the eluent and Tc-99m enter."

    You see it's TRIVIALLY EASY to separate the Tc-99m from the Mo-99.
    It's done merely by passing an eluent over the Mo-99. That's hardly "impractical"
    as you stated above, and doesn't require the hospital to be anywhere near the

    What would be impractical is to separate out Tc-99m directly and ship that.
    The Tc-99m isotope is short-lived - you don't want to be giving long lived
    radionuclides to patients. So a hospital would need an essentially constant
    stream of Tc-99m replenishing their supply.

    No - instead they are given the Mo-99 precursor to Tc-99m. That way they
    have a constant supply of Tc-99m being made for them by radioactive decay
    of the longer lived Mo-99. That Mo-99 comes from targets that are irradiated
    in research reactors. As the above linked article states, the Mo-99 "cow" is
    replaced weekly.

    Sure that's "inefficient" in that there are Tc-99m atoms that decay without
    ever getting used - but SO WHAT!!!

    Dr. Gregory Greenman
    Last edited by a moderator: May 2, 2017
  17. Dec 29, 2006 #16
    http://www.iaea.org/OurWork/ST/NE/NEFW/nfcms_researchreactors_Mo99.html" [Broken]

    Perhaps in research facilities it is true that neutron activation technique is used, but most Mo-99 used in medical facilities and nuclear pharmacies in the US comes from Mallinckrodt and BMS. I believe BMS gets their Moly from Nordion http://www.nordion.net/documents/elibrary/molecular-isotopes/MO-99/Mo-99_Can.pdf" which uses a 235U(n,f) reaction. The other reactor that produces a great deal is a consortium in Brussels (I believe) that has a backup reactor in South Africa. There is no production of Mo-99 in the US (on a large scale, that is). It all comes from outside the country, hence no problems with reprocessing spent fuel. Also,
    from http://www.nuclearonline.org/PI/Nycomed%20Mo%2099-Tc%2099m%20Genera.pdf" [Broken]
    Last edited by a moderator: May 2, 2017
  18. Dec 29, 2006 #17
    My bad. I meant to say products of fission products. Or products of products of fission products (not sure how some of the radionuclides used in nuclear medicine are produced). Since Tc-99m is the most widely used radiopharmaceutical, I would think my use of the word "most" is still applicable, though. I'm not sure how I-131 is captured in the fission process, or whether that is indeed an activation product (my guess is it's easier to use the fission product I-131).
    I never said you want something isotopically pure for the patient, but i do agree you want the parent nuclide (Mo-99 in this case) to be as isotopically pure as possible. Problem is, as I've said, it can't be done.

    Well, USP would disagree with that, but I understand what you mean. USP says there is a minimum% (between 80% to 95%, depending on which drug the technetium is tagged with) of technetium that is tagged to the drug in question that must be Tc-99m.
  19. Dec 29, 2006 #18


    User Avatar
    Science Advisor
    Dearly Missed


    If we're talking about suppliers outside the USA; then yes.

    They don't have the isotopic separation technology that the USA has.
    Isotopic separation is a sensitive technology, because it can be used to
    make HEU - highly enriched uranium.

    However, if one has isotopic separation technology, it's better to prepare
    a precursor for irradiation rather than scavenging from fission products.

    Dr. Gregory Greenman
    Last edited by a moderator: Apr 22, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook