Polarization : A Must for Fusion

Recent realistic simulations confirm that the polarization of the fuel would improve significantly the DT fusion efficiency. We have proposed an experiment to test the persistence of the polarization in a fusion process, using a terawatt laser hitting a polarized HD target. The polarized deuterons heated in the plasma induced by the laser can fuse producing a He and a neutron in the final state. The angular distribution of the neutrons and the change in the corresponding total cross section are related to the polarization persistence. The experimental polarization of DT fuel is a technological challenge. Possible paths for Magnetic Confinement Fusion (MCF) and for Inertial Confinement Fusion (ICF) are reviewed. For MCF, polarized gas can be used. For ICF, cryogenic targets are required. We consider both, the polarization of gas and the polarization of solid DT, emphasizing the Dynamic Nuclear polarization (DNP) of HD


Introduction
The polarization of D and T nuclei should increase their reactivity when used as fuel material in fusion processes induced either by magnetic or by inertial confinement.The fusion reaction: + n + 17.6 MeV (1) goes mainly through the excitation of an 5 He 3/2 + intermediate state, resulting from the coupling of the spins 1 and 1/2 of the D and T nuclei to a total spin S = 3/2.Without polarization of D and T, the statistical distribution of the six possible states gives four S = 3/2 and two S = 1/2 states.Only the 3/2 states can produce the intermediate 3/2 resonance.With 100% parallel polarization of D and T, all states would contribute to the fusion, increasing the reactivity by 50%.In addition, the polarization allows the control of the direction in which the reaction products are emitted, the neutron having a sin 2 distribution.This can be very useful to reduce damage or activation of costly equipment.Theoretical considerations have indicated that the polarization should persist as well in MCF [1] as in ICF [2].In addition, recent realistic simulations show that the required hot-spot temperature (T) and areal density ( R) can be reduced by about 15 % for fully polarized nuclear fuel.and energy ratio E B / E A as a function of gain G (dashed line).Subscripts A (blue) and B (red) refer respectively to un-polarized and polarized DT fuel.Moreover, numerical simulations of a directly driven capsule show that the required laser power and energy to achieve high gain are significantly reduced, while the maximum achievable energy gain scales roughly as the fusion cross section [3]. Figure 1 from [3] summarizes those results.It is found that for an invested energy of 1 MJ, the estimated gains are G A (1MJ) = 130 while G B (1MJ) = 300 for un-polarized and polarized DT fuel respectively.Moreover, for a fixed gain G, the ratio E B / E A is always smaller than 0.7, decreasing as the gain increases (see the dashed line in figure 1) and indicating a decrease of the required invested energy E B for polarized fuel.As an example, for a gain G = 100 the driver energy is E A = 880 kJ while a smaller energy E B = 510 kJ (E B / E A = 0.58) is required.Obviously, very high gains can only be achieved with polarized DT fuel.It should be noted that the sin 2 distribution of the polarized reaction products was not taken into account in the above simulation.A narrower concentration of the heat due to such an anisotropic distribution should further favour the generation of hot spots.

Persistence of the polarization
We have proposed to investigate the polarization persistency using the reaction: induced by fusion of polarized deuterons heated in a plasma.It is anticipated that the angular distributions of the neutrons as well as significant changes in the fusion rates can be measured and related to the persistence of the polarization.Details on the setup and the experiment features are given in Ref. [4].Here, we show only a sketch of the experiment.

Polarization of Hydrogen isotopes
The experimental polarization of DT fuel is a technological challenge.For MCF, polarized gas can be used.Atomic beams of H and D are currently produced with high polarization, but their intensity is too low by 3 orders of magnitude [9].It has been proposed to polarize Hydrogen molecules, which allows the production of molecular beams with a satisfactory intensity of 10 20 molecules/s [10].This is barely enough to feed ITER.For ICF, cryogenic targets are required.The symmetries of H 2 , D 2 or T 2 Hydrogen isotopes in their homo-molecular form (two fermions or two bosons), trapped as long lived meta-stable states in the solid, complicate the polarization of corresponding targets.However H D, a hetero-molecular form of Hydrogen has no symmetry and can be polarized [11].The static polarization (high field and low temperature) using distilled H D has been achieved [5], opening the possibility to produce polarized DT molecules [12].High static polarization of DT cannot work, because of the heat generated by the Tritium radioactivity (~ 640 mW/g).Therefore, one has to presents specific additional difficulties [13].Still, the DNP of H D itself was investigated in the past with only moderate success [14].Therefore, although quite possible in principle, the polarization of DT fuel will require huge technological developments.Figure 3 shows the standard scheme of DNP polarization of H D or DT: the high static polarization of electrons is transferred to H or T nuclei by RF transitions (see [12] for more detail).For efficient DNP, long relaxation times T 1 H and T 1 D for the nuclei and short relaxation times T 1 e for the electrons are required.This is normally the case for distilled H D [5 optimum DNP polarization of H D molecules that could possibly be applied to DT molecules which have the same magnetic and molecular configurations as H D. The moderate success of Solem should not be discouraging.He has used commercial H D containing several percents of H 2 and D 2 , which are trapped as ortho-H 2 (o-H 2 ) or para-D 2 (p-D 2 ) long living metastable states in the solid H D. Those magnetically actives species reduce the nuclear T 1 relaxation times and limit the DNP process.Highly distilled H D or DT, with impurities contaminations less than 10 -4 should be used [5].It has been pointed out that the energetic electrons due to the radioactivity of Tritium would dissociate DT molecules allowing the D and T ions to recombine in D 2 and T 2 poisoning quickly any purified DT gas [13].However, the equilibrium configurations of those Hydrogen isotopes at low temperature are o-D 2 and p-T 2 which do not allow thermal connection between the lattice and the spin reservoir and have therefore little impact on the T 1 spin-lattice relaxation times.As a benefit, the huge abundance of free electrons in DT should be a big help for the DNP process.Another difficulty is that the e-DNP (DNP from electrons) of D nuclei gives rise to low D polarizations of the order of 40%, while the polarization of H nuclei (protons) can approach 100 %.Therefore, the n-DNP (DNP from nucleons) should be used to increase the D polarization.This process called Adiabatic Fast Passage is used to increase the D polarization in polarized HD targets by transferring the H polarization to D [15].The H polarization is lost, but can be regenerated by e-DNP and the whole process repeated several times to reach high H and D polarizations [16].In US, there is a project to inject 55% polarized D ( from HD molecules) and polarized 3 He in the DIII-D Tokamak of San Diego, in order to see a 15% increase of the emitted protons by the fusion reaction: D + 3  4 He + p + 18.35 MeV due to the fuel polarization [17].
The relatively high D polarization of 55% would be obtained by a single n-DNP process at the expense of the H polarization, after static polarization of the solid HD [18].Finally, one should note that solid polarized targets can be injected in MCF reactors, as proposed in [17].

Conclusion
A considerable effort is under way to produce energy using controlled fusion either by MCF or ICF.Polarized fusion fuel is of great interest, both to increase the fuel reactivity and to control the direction in which reaction products are emitted.The question to know if the polarization will persist in a fusion process can be answered using existing experimental equipment, and neutron detection.The corresponding investigation is underway at ILE (Osaka).The ongoing measurements at Gatchina for the double polarized D D fusion at low energy will provide the necessary experimental data to analyze the Osaka results.Finally, the polarization of the DT fuel is a huge problem, but the impact of the polarization on fusion is so important that a revival of the field is absolutely needed.
The successes with the static polarization of H D allow big hopes for the DNP of H D and DT.The production of high intensity polarized molecular beams is another way to pursue.
count on the Dynamic Nuclear Polarization (DNP) of solid DT which, compared to the DNP of H D,

F ig. 3 .
Scheme showing the DNP polarization of Protons or Tritons.