Introduction

            Nuclear Magnetic resonance (NMR) is one of the most powerful analytical techniques in modern chemistry and biological sciences. It is used to identify and characterize molecular structures, providing unique information about inter- and intramolecular interactions, molecular dynamics, hybridization, or strength and orientation of chemical bonds. It also finds routine use in noninvasive imaging of the interior of human body, providing medicine with one of the most effective diagnostic tools – the technique of Magnetic Resonance Imaging (MRI). Besides the well-known medical applications, MRI is also used in other disciplines ranging from prospecting for natural resources to food industry.

            NMR experiments are usually performed at high magnetic fields (larger than 1 T). On one hand, application of such fields is necessary to polarize/prepare samples and on the other it is dictated by a conventional detection schemes, where inductive coils are used to measure NMR signals. Application of fields several hundred thousand times larger than the Earth magnetic field provides NMR with chemical sensitivity and offers good spatial resolution of MRI, but also brings about various technical challenges. Particularly, generation of the fields requires application of cryogenically-cooled superconducting magnets (operating at 4 K), that require liquid helium. Additional challenge is the requirement of extremally high spatial homogeneity of the field (10^-6 or smaller). This dramatically increases the price of NMR/MRI devices and their maintenance. The high fields also introduce limitations regarding application of the method, making MRI unsuitable for patients with pacemakers or metallic implants/shrapnels and NMR for detection of non-magnetic samples. Finally, the requirement of magnetic field homogeneity severely limits the mobility of both chemical and medical NMR/MRI machines.

            Our aim is to develop a new approach to NMR. Its most important feature is operation at ultra-low magnetic fields (i.e., fields significantly lower than the Earth’s magnetic field), or even without any external fields whatsoever. Operation at such novel experimental conditions provides NMR with new capabilities, opening fields traditionally unavailable to the technique.

Technical challenges of zero and ultra-low nuclear magnetic resonance

            To conduct NMR experiments at zero or ultra-low magnetic fields (ZULF), several technical obstacles need to be overcome. One of them is polarization/preparation of the sample at ZULF. Since thermal polarization, i.e., the technique routinely used for polarization of NMR samples, scales linearly with the field strength, operation at ZULF does not provide enough polarization for NMR signals to be detected by conventional means. This, however, may be addressed by sample prepolarization, where the sample is placed in a strong magnetic field prior to the measurements. Since the polarization process does not require homogeneous field, the it may be realized using permanent magnets, which are significantly cheaper than their superconducting counterparts, and allow us to dispense with the cryogenic cooling of the apparatus.

            As mentioned, prepolarization can be performed outside of the NMR-signal detection region (i.e. remote polarization). In such a case, the sample needs to be mechanically shuttled between the polarization and detection regions in the timeframe significantly shorter than its longitudinal relaxation time. An alternative to this approach are the hyperpolarization techniques such as optical pumping, dynamic nuclear polarization, parahydrogen induced polarization. These processes enhance sample’s polarization by some orders of magnitude with respect to what can be achieved by thermal polarization even at the strongest fields generated with superconducting magnets. This in turn leads to NMR signal amplitudes being improved by many orders of magnitude. Moreover, thanks to the fact that some of the hyperpolarization techniques do not require any magnetic fields at all one can perform magnet-free NMR.

            Another problem of NMR at ZULF is signal detection. Conventionally, NMR signals are detected with inductive coils, which measure oscillating magnetic fields generated by spins evolving in strong external magnetic field. Since the voltage induced in the coils scales linearly with the speed of spins’ precession, the stronger the field, the better the detection sensitivity of the technique. Importantly, the field needs to be extremely homogenous to not to broaden the observed NMR lines, and thus corrupt the sensitivity of the method. At low fields the magnetic (i.e. Larmor) precession is slow and the technique suffers from a dramatic loss of sensitivity up to the point where no NMR signals can be detected using the scheme. Therefore, to measure slowly oscillating NMR signals (<1 kHz), we substitute the inductive detection scheme with the most sensitive magnetic-field sensors yet created – optical magnetometers (sensitivities on the order 1 fT/Hz^1/2 or better). These sensors exploit the so-called magneto-optical phenomena, in which light-matter interaction is affected by the presence and parameters of the magnetic field. This leads to a change of polarization and/or intensity of light propagating though the magneto-optically-active medium. Optical magnetometers are free of the limitations of conventional inductive coils and enable detection of NMR signals not only at low, but also truly zero external magnetic fields. This approach opens many possibilities and has already demonstrated its applicability in ZULF NMR. The remaining question is what sort of information we can gain at ZULF.

J spectroscopy at ZULF

            A crucial feature of NMR is its ability to extract information about the chemical structure of molecules. This is thanks to the so-called chemical shift, the effect stemming from the slight change in local magnetic field at the point of the nucleus due to surrounding electrons (the effect is called diamagnetic shielding). The shift scales linearly with the magnetic-field strength, and without the field it disappears making the technique unusable. However, there is another phenomenon that becomes manifest only at extremely low fields that can be used for extraction of structural information. It is the so-called scalar J-coupling that exists between spins of atomic nuclei in molecules. This interaction, mediated by the molecule’s electrons over distances of several chemical bonds, provides information about molecule’s spatial structure, and the strength and orientation of its chemical bonds.

Applications of ZULF NMR

            At ZULF the relaxation of sample magnetization, including the dominating dipole-dipole relaxation, is slowed down. This translates to very long lifetimes of sample’s polarization, reaching several tens of seconds. This means that the spectral lines observed in ZULF NMR corresponding to interaction between different atoms in the molecules, are extremely narrow (sub-hertz). Given that they are also unique for a given compound, two potential areas of application for ZULF NMR open. The first being chemical fingerprinting – enabling recognition of unknown substances. In the long run, equipped with a library of zero-field NMR spectra and dedicated software, this would allow identification of e.g. liquid explosives at security gates at the airport. Apart from chemical identification, the ultra-narrow linewidths can be used to systematically measure the influence of external factors (like pH and temperature) on molecular structure.

            On the other hand, the ability to investigate minute changes of NMR signals induced by internal or external factors. This can be performed at an unprecedented level, providing extremely sensitive physico-chemical sensors, for metrology and other fundamental research like e.g. searches of exotic interactions that were proposed theoretically, but never observed experimentally. This will get us one step closer to understanding some of the most profound questions of modern science including what the Dark Matter and Dark Energy is, and why is there so much more matter than antimatter in the observed universe.

            Due to the simplicity of achieving better absolute homogeneity of the magnetic field than at classically used strong fields, the low-field Magnetic Resonance Imaging will provide significantly higher image resolution. Thanks to the slowly varying magnetic field pulses used in low-field MRI being shielded to a lesser degree by the skin effect, it will be possible to image metallic samples, and samples inside metallic enclosures. Finally, the apparatus can in principle be miniaturized to such an extent, that creating a mobile NMR laboratory, or NMR laboratories small enough to be a convenient addition to any medical clinic will become a reality.

Conclusion

            Many of the above-mentioned schemes and applications of ZULF NMR/MRI are being currently developed in our laboratory in Krakow. To obtain more information see our recent publications.

www.zulf.eu

[1] Xu, Shoujun, et al. "Magnetic resonance imaging with an optical atomic magnetometer." Proceedings of the National Academy of Sciences 103.34 (2006): 12668-12671.