I am looking to achieve isotope separation using transition states.

In the rxn of dienes with halogens, based on the temperature, the dienes can create a thermodynamic product or a kinetic product. The kinetic product is reversible, while the thermodynamic product is not. This is because thermodynamic products have a higher energy need for transition states.

In the context of transition states, I see some possibility of having two different products, each having one type of isotope, by selectively exciting one isotope while leaving the other isotopes untouched.

I look to achieve this by the following protocol:

  1. Dissolve the isotopes in some liquid.
  2. Put the liquid-isotope solution in NMR machine.
  3. Make NMR machine "tip over" only one variety of isotope repeatedly to gain thermal energy.
  4. Maintain temperature of liquid so that other isotopes not tipped over can only maintain the kinetic product.
  5. Make NMR machine energize target isotope enough so that it can produce thermodynamic product.
  6. Put in the diene to react with isotopes.
  7. Seperate the now permanent thermodynamic product, with some pre-existing method to separate isomers.
  8. Yield the target isotope.

Is this method viable? I am proposing this method with the assumption that thermodynamic and kinetic products can be produced based on the individual energies of halogens, which I am not entirely sure of.

In addition, would this mechanism be viable for reactants other than halogens (using a different TS and rxn, of course)? I am interested in enriching heavier metals, such as uranium (where conveniently, 235 reacts with NMR while 238 does not).

Thank you.


2 Answers 2


There are at least two glaring problems:

  1. Look at the energy scales on which NMR operates (for example, take the resonance frequency $\nu$ and calculate the associated energy $h\nu$)... and compare it to, for example, the thermal energy $k_\mathrm{B}T$. You'll find that the energy absorbed by nuclear spins in a typical NMR experiment is incredibly negligible. We're talking several orders of magnitude too small.

  2. Even if you can somehow put in enough energy via NMR excitation and convert it into thermal (kinetic) energy, this energy will be dissipated amongst every molecule in your reaction sample. You can't have some molecules in your solution be at a higher temperature than the rest, not in a meaningful manner, at least.


Look up the difference of the isotopic masses between hydrogen and deuterium (a factor of about 2),* and the two isotopes of uranium (difference here $\approx1.5\%$), e.g. in NIST's table of atomic weights. From this, and the examples in Wikipedia's article about the kinetic isotopic effect where the effect correlates with difference of the isotopic masses concerned, you may infer for reactions evolving uranium at conditions typically seen in the lab, the effect likely will be too low to be of practical value.

Similar to orthocresol's argument, NMR spectrometers excite samples with energy in the range of radio frequency. The energy per photon of this radiation is too small to excite a molecule beyond the typical thermodynamic, or photochemical activation threshold of a reaction. However, NMR kinetic studies can vary the temperatures of samples to accelerate / slow reactions (keyword Arrhenius equation), or to render some reaction pathways practically inaccessible (barrier of activation energy). Practically, because in perspective of statistical thermodynamics, not all molecules in a sample carry the same energy. Instead, there is some tailed (Boltzman) distribution of energy among the molecules.

* Use of deuterium in chemicals and pharmaceuticals has its place, though. See e.g., the youtube video How "Heavy Hydrogen" makes Drugs work: Deuterium in Pharmaceuticals, Organic Chemistry & Synthesis

  • 1
    $\begingroup$ @YoungJunLee If you have an afternoon: Reaction Monitoring by NMR getting the most from the data Webinar as a tutorial (to replicate, you would need both their basic software + a separate plug-in), or Time- and site-resolved kinetic NMR for real-time monitoring of off-equilibrium reactions by 2D spectrotemporal correlations in 2022NatureCommun833 as an open access publication to showcase an application. (Work with 2D NMR spectra requires an other plug-in.) $\endgroup$
    – Buttonwood
    May 11, 2023 at 7:18

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.