Long nuclear spin coherence times for molecules trapped in high-purity solid parahydrogen

TL;DR

The paper addresses achieving long nuclear-spin coherence times for molecules trapped in a solid matrix to enhance precision measurements of symmetry-violating physics. The authors trap HD in high-purity solid parahydrogen and control the orthohydrogen impurity fraction $X$ via an inline catalyst, then measure $T_2^*$, $T_2$, and $T_1$ as a function of $X$. Key findings show that in the dilute impurity regime both $T_2^*$ and $T_2$ increase roughly as $1/X$, with a low-$X$ plateau for $T_2$ around $0.3$ s, and $T_1$ scales approximately as $X^{-2}$ within a similar range; the proton–deuteron coupling is resolved as $J(H,D)=47.2 \pm 1.1$ Hz. The work demonstrates a scalable solid-matrix platform with long nuclear-spin coherence and outlines strategies for polarization restoration and future studies on other diamagnetic molecules to probe physics beyond the Standard Model.

Abstract

We measure the ensemble transverse relaxation time (T2*) and spin-echo coherence time (T2) of the proton spin of HD molecules trapped in solid parahydrogen. By using high-purity parahydrogen matrices, we are able to measure significantly longer T2 and T2* times than seen in prior work. We also measure the longitudinal spin relaxation time T1. We examine how these parameters scale with the matrix purity and find limits on the coherence time from the parahydrogen matrix itself.

Figures (5)

• Figure 1: Scale-accurate schematic of the apparatus, as described in the text. The saddle coil and support tube are shown in cutaway views, and the cylindrically-symmetric magnet is shown in cross-section.
• Figure 2: Power spectrum of the FID signal for two samples of differing orthohydrogen fraction $X$, as labeled. The broad and narrow lines have a frequency offset $f_0$ of 58.41 and 56.87 MHz respectively.
• Figure 3: The spin-echo $T_2$ --- plotted as $1/T_2$, the decoherence rate --- as a function of the relative HD flow rate: the ratio of the gas flow rates of HD and parahydrogen during sample growth. Data is shown for four samples, all with an orthohydrogen fraction of $1 \times 10^{-4}$, and for three different spin echo sequences. The lines are a linear fit to the data. We note the HD fraction in the sample is smaller than the relative HD flow rate (based on NMR FID amplitudes).
• Figure 4: $T_2^*$ and $T_2$ as a function of the orthohydrogen fraction $X$, as discussed in the text. The dashed lines are proportional to $X^{-1}$ and are included as a guide to the eye. The orthohydrogen fraction is determined from the mean temperature of the catalyst during deposition doi:10.1063/5.0049006. The catalyst temperature varies during deposition, with typical standard deviations of $\sim 2\%$.
• Figure 5: $T_1$ as a function of the orthohydrogen fraction $X$. At high values of $X$, the HD naturally present in the hydrogen is sufficient that additional doping with HD is not needed; undoped samples are shown as black points. For $X \lesssim 10^{-3}$, the catalyst suppresses the HD and no signal can be observed without additional doping. The dashed line is proportional to $X^{-2}$ and is included as a guide to the eye.