The hyperpolarization of nuclear spins like 13C is a promising approach to access the hidden orders of magnitude of signal in MR imaging and spectroscopy, holding great potential to improve medical diagnostics (1,2). Parahydrogen‑based methods employ the inherent spin order of parahydrogen (pH2) and critically rely on its purity (3-7). For maximum yield, an enrichment of as close to 100 % is demanded prior to any hyperpolarization experiment.
pH2 is the spin-singlet isomer of dihydrogen (Fig. 1), which has four spin states in total (one para, three ortho (oH2)). Following the Boltzmann distribution, these four states are approx. equally populated at room temperature (rtH2), i.e. rtH2 = 25 % pH2 and 75 % oH2. At lower temperatures, the equilibrium is shifted towards the para-isomer, reaching a pH2 enrichment fraction (fpH2) ~ 100 % close to the boiling point of H2 (20 K). While the mechanism of interconversion between both fractions is still being discussed (8,9), the life time of para-enriched hydrogen generally exceeds hours or days, dependent on its interactions with the storage container.
The spin properties of H2 have been the subject of discussion since the early days of quantum mechanics (10-12) and several methods were suggested for the production of parahydrogen for various applications (e.g. (13-15)). All rely on contact with a catalyst at low temperatures (e.g. liquid N2 at 77 K, fpH2(77 K) ~ 50 %). Recently, some experimental setups have been described aiming at the production of pH2 for hyperpolarization (16,17). Generally, pH2 production is not trivial as H2 poses a considerable safety risk, being highly explosive when mixed with air at 4 % ‑ 76 %, and aggressive to some materials (causes hydrogen embrittlement). Furthermore, the previously described designs and procedures are (a) relatively low in production rate (less 1 standard liter per minute, ls/m, SI units), (b) low in pressure (less 10 bar), (c) low in enrichment (50 %) or (d) can’t be translated easily to a hospital (e.g. access to an NMR spectrometer, or continuous scan times exceeding 50 hours). These may be the reasons why the implementation and operation of a production unit in a clinical setting, to our knowledge, has not been attempted yet.
In the scope of hyperpolarization’s application in life sciences, however, this is a prerequisite: future applications are likely to be found in the vicinity of medical research, e.g. in hospitals, where the production of pH2 is severely restricted. It is the aim of this contribution to lower this hurdle, which represents but one on the way to routine clinical use of pH2 hyperpolarization.
We developed a high-throughput, high-pressure continuous-flow parahydrogen converter exceeding the specification of its predecessors (Fig. 2). Furthermore, we present its safety concept, quality assurance and routine operation protocols with the goal to supply pH2 safely for subsequent biomedical hyperpolarization research in medical institutions.

Am Botanischen Garten 14
24118 Kiel • Germany

Tel. +49 (0)431 880 58 32

How to find us


Go to top