In NMR a spin is taken to an excited state under a magnetic field and then it "relaxes spontaneously" to the original state emitting the signal that is measured. In fact the relaxation is not spontaneous, the spin needs certain stimulation to achieve it, we know that from the quantum transition theory. In practice the stimulation comes from random photons flowing through the sample. My question is then whether is possible to stimulate the sample and make it relax very quickly using photons with the Larmor frequency. The situation is analogous to that of a laser. One of the most important applications of this method would be the shortening of measurement times and the increase of sensibility in NMR experiments including MRI
UPDATE
I saw this in Physics Today
http://scitation.aip.org/content/aip/magazine/physicstoday/article/69/3/10.1063/PT.3.3097?utm_source=Physics%20Today&utm_medium=email&utm_campaign=6832173_March%202016%20Table%20of%20Contents&dm_i=1Y69,42FQL,JNYN94,EQDPP,1
Seemingly, Purcell himself had a similar idea in 1946
Ok, perhaps the photons won't have the Larmor frequency, I'm not sure right now, but this doesn't affect the general idea.
The idea of having "photons at the Larmor frequency" sound to me rather like the RF excitation of spin transitions (either pulsed or continuous-wave) that is done during an NMR experiment. Involving the absorption and excited emission of photons, such an excitation leads to the saturation of the nuclear spin transitions, i.e. the equilibration of the associated energy levels.
The source of nuclear spin relaxation towards thermal equilibrium are random fluctuations of the spin Hamiltonian, e.g. the dipolar interaction of nuclear spins on a molecule that undergoes rotational diffusion in a liquid or phonons in solids. The theoretical treatment of nuclear spin relaxation using perturbation theory (Bloembergen/Purcell/Pound or Bloch/Wangsness/Redfield) tells that the nuclear spin relaxation rates depend on the spectral density of these fluctuations at the nuclear spin Larmor frequency. Roughly, if the time scale of these fluctuations (characterized by their correlation time) is equal to the inverse of the nuclear spin Larmor frequency, relaxation will be most efficient. To summarize, nuclear spin relaxation can be enhanced if randomly fluctuating spin interactions with strong spectral components at the nuclear spin Larmor frequency are introduced to the system. Another way to look at these relaxation processes is the concept of spin temperature. Basically, stimulating the nuclear spin system by an RF pulse, as its done in most NMR experiments, transfers energy to the nuclear spin system, increasing its temperature. In order to relax towards thermal equilibrium, it needs to be coupled to a thermal bath that can take the energy introduced by the RF irradiation. This coupling may be a dipolar coupling among two nuclear spins and the thermal bath would then be the random motions of their molecules.
Hence, as far as I know, your idea would not work. It should only introduce energy into the system (further increasing the spin temperature) instead of cooling the system to reach thermal equilibrium.
And how is different from laser? In laser there is a population inversion in which the atoms adquire an state away from equilibrium and you make them get to their ground state by photon stimulation
Isn't it that the probability of stimulated emission and absorption is equal and the overall effect of bringing atoms to their ground state only results from the continuous pumping and population inversion? There should be no overall change of population once both levels have equal population and you should not be able to reach the thermal equilibrium distribution of atoms in the ground and excited state by irradiation of light.
Yes, this works; you can stimulate the return to the ground state, or bypass the T1 relaxation time. To my understanding this is the same phenomenon as stimulated emission with a laser. But if you rapidly excite- and relax- your sample (for example, a 90° pulse followed immediately by a -90° pulse) your sample won't really be at equilibrium because the phase coherences will still be present (owing to the non-zero T2 relaxation time) even if the entire population is equally excited as the equilibrium case.
Armando,
May I suggest that you examine the papers by Graham Fleming. His group at Berkeley has done extensive work with measuring photon echos and dephasing times.
Best,
John
Agreeing with answers already posted, using photons at the Larmor frequency, e.g. a 90 degree pulse, would saturate the system, i.e. equalize the populations in the spin sublevels, raising the spin temperature to infinity. But it would leave a coherent component in the transverse plane, so the system would not be completely relaxed. And anyways, it would not be at the temperature of the bath. Adding paramagnetic impurities could speed the relaxation (to thermal equilibrium) but this would not be a method using "photons at the Larmor frequency".
..hmm...to my remembering (aquired a loooong time ago) a laser is a 3-level system. Wherby the high energy level is pumped from a groundstate via irradiation, this state relaxes to a mid-energy state quickly. The transition from this mid energy state to groundstate is forbidden - therefore the population of this state becomes high. Finally the back-transfer to the ground state can be triggered (stimulated emission) by photons of suited energy....
- Badges
- Science topic
- Similar topics
- Biological Science
- Histology
- Fixatives