Solid state 13C dynamic nuclear polarization (DNP) experiments were conducted on a homebuilt 7 T quasi-optical DNP polarizer, on static samples of [1-13C]pyruvic acid and glycerol with 15 mM OX063 trityl radicals. The effects of the addition of Gd-chelate Gd595 and the absence of glycerol on the DNP and electron paramagnetic resonance (EPR) spin dynamics are compared and contrasted to draw important conclusions on main DNP contributors. DNP and EPR experiments, at 7 T (197 GHz) and 8.5 T (240 GHz), respectively, were conducted in a dynamic temperature range of 35 K to 3.7 K where the electron thermal polarization changes from 13.4 % to 85.6 %, respectively. Inhomogeneously broadened trityl EPR lines are found and the dominant DNP mechanism is shown to be the cross effect (CE). We have achieved 20 % 13C polarization at 3.7 K and 7 T with a [1-13C]pyruvic/glycerol/H2O sample (p-Mixed) and the effect of 2 mM Gd595 drops the 13C polarization to 11% (p-Gd-Mixed).

Only a 6.77% 13C polarization was found with the more highly 13C concentrated pure [1- 13C]pyruvic acid sample (p-Pure). For all samples, EPR linewidths and T1e are heavily temperature dependent in the range 3.7-35 K. T1e values, for all samples, at 8.5 T seem to follow the direct process and are much smaller than what has been reported in literature[1-4] for identical systems at lower magnetic fields. The electron phase memory time (TM) doesn't seem to have much effect on the buildup of nuclear polarization (TDNP) or the DNP enhancement but that the two strongest factors in determining equilibrium DNP enhancements are T1e and T 1n. However, when T1n becomes sufficiently long T1e seems to be dominating the DNP process.

The nuclear spin lattice relaxation (T1n) seems to fit well with theory when it is assumed that T 1n relaxation occurs through fixed paramagnetic impurities and is modulated by the electron-electron dipolar reservoir (∼TM) as the correlation time of the electronic field that the nuclei "see". It is shown that, in some cases, longer T1e values seem to correlate to larger DNP enhancements, but some literature studies contradict this apparent observation. In fact, the Borghini model for low temperature DNP predicts that a shorter T 1e is favorable for higher DNP enhancements. Overall, the Borghini model for low temperature DNP provides a qualitative validation of the main factors that influence DNP enhancement as well as the temperature dependence of DNP enhancements.