1. Introduction

1.1. Motivation

Nuclear magnetic resonance spectroscopy is a powerful tool for the investigation of material properties, which has found applications in many areas of physics, chemistry, biology and medicine. Its physical foundations, the excitation of transitions between nuclear spin states, was explored in the years after the second world war, and is well characterised by now. Today's interest in the field is based largely on the immense potential for applications: spins can serve as probes for their environment because they are weakly coupled to other degrees of freedom. In most magnetic resonance experiments, these couplings are used to monitor the environment of the nuclei, like spatial structure or molecular dynamics.
While the direct excitation of nuclear spin transitions requires radio frequency irradiation, it is often possible to use light for polarizing the spin system or for observing its dynamics. This possibility arises from the coupling of spins with the electronic degrees of freedom: optical photons excite transitions between states that differ both in electronic excitation energy as well as in their angular momentum states. In general, the light can serve three different purposes: it can polarize the spin system, thereby creating the 'raw material' for spectroscopic experiments; it can measure the spin polarization, serving as a detector; and it can influence the dynamics of the system. Some motivations for using light in magnetic resonance experiments include

Sensitivity: In many cases, the possible sensitivity gains are the primary reason for using optical methods. Compared to NMR, sensitivity gains of more than 10 orders of magnitude are possible. In fact, is was recently demonstrated, that it is possible to measure magnetic resonance spectra of individual molecules [Koehler et. al. 1993; Wrachtrup et. al. 1993], while a typical NMR sample contains of the order of 10¹⁷ spins.
Selectivity: Lasers can be used to selectively observe signals from certain regions in space, like surfaces, at certain times which may be defined by laser pulses with a resolution of 10^-14 sec, or from a particular chemical environment defined, e.g. by the chromophore of a molecule or the quantum confined electrons in a semiconductor. Crystal imperfections can change the optical properties of an atomic or molecular site in such a way that it those sites can be distinguished against a background that is orders of magnitude larger.
Speed: Conventional magnetic resonance requires the presence of a population differences between spin states to excite transitions between them. This population difference is usually established by thermal relaxation through coupling with the lattice, i.e. the spatial degrees of freedom of the system. At low temperatures, this coupling process may be too slow for magnetic resonance experiments. In the case of optical excitation, the population differences are established directly by a polarizing laser beam. The time required for polarization of the system depends then only on the laser intensity and can be instantaneous on the time scale of the magnetic resonance experiment, independent of temperature.
Experiments in electronically excited states: If information about an electronically excited state is desired that is not populated in thermal equilibrium, it may by necessary to use light to populate this state. It is then certainly advantageous to populate the different spin states unequally to obtain at the same time the polarization differences that are needed to excite and observe spin transitions.

1.2. Laser Spectroscopy

Apart from magnetic resonance, the interaction of matter with laser light is the second most important ingredient for the experiments that will be discussed below. In laser spectroscopy, light induces transitions between different electronic states. The differences between laser spectroscopy and magnetic resonance are related to the different frequency scales: optical frequencies are more than 6 orders of magnitude higher than rf frequencies. An important consequence is that laser spectroscopy is much more sensitive as a technique: optical experiments on individual atoms [Hurst et. al. 1979; Dehmelt 1990], and molecules [Nguyen et. al. 1987; Moerner et. al. 1989] have become routine in the last few years. While the first experiments of this type were performed in ultrahigh vacuum, they can now also be performed in condensed matter, like solutions [Nguyen et al. 1987] or solid matrices [Moerner et al. 1989].
Another characteristic property of laser spectroscopy is the high time resolution which is of course also intimately linked to the high frequencies of optical radiation. Technical limits to the time resolution are now of the order of a few femtoseconds, corresponding to only a few optical cycles. These characteristics are of course used extensively in various applications. In chemistry, the high time resolution has been used for 'real time' monitoring of chemical reactions, e.g. by measuring the intermolecular potential as a function of time during the reaction. Also in solid state physics, laser spectroscopy is used extensively to monitor fast processes, like excitation and recombination of electrons and holes in semiconductors.
In the field of high-resolution spectroscopy, it is also possible to achieve very high frequency resolution: some systems can now reach resolutions of better than 1 Hz, corresponding to a relative frequency stability of some 10¹⁴. This high frequency resolution is especially attractive for metrology, where many groups try to build new, more precise frequency standards that are based on optical transitions. The properties of laser spectroscopy, high frequency resolution and high temporal resolution can potentially also be used in optically enhanced magnetic resonance experiments - obviously an attractive perspective for many applications.
The article is structured as follows: the next section discusses the most important physical phenomena that permit the use of laser radiation in magnetic resonance experiments. Examples of applications to specific systems are discussed in the third section.

References

H. Dehmelt, 'Experiments with an isolated subatomic particle at rest', Rev. Mod. Phys. 62, 525-530 (1990).
G.S. Hurst, M.G. Payne, S.D. Kramer, and J.P. Young, 'Resonance ionization spectroscopy and one-atom detection', Rev. Mod. Phys. 51, 767-819 (1979).
J. Koehler, J.A.J.M. Disselhorst, M.C.J.M. Donckers, E.J.J. Groenen, J. Schmidt, and W.E. Moerner, 'Magnetic resonance detection of a single molecular spin', Nature 363, 242-243 (1993).
W.E. Moerner, and L. Kador, 'Optical detection and spectroscopy of single molecules in a solid', Phys. Rev. Lett. 62, 2535-2538 (1989).
D.C. Nguyen, R.A. Keller, and M. Trkula, 'Ultrasensitive laser-induced fluorescence detection in hydodynamically focused flows', J. Opt. Soc. Am. B 4, 138-143 (1987).
J. Wrachtrup, C.v. Borczyskowski, J. Bernard, M. Orrit, and R. Brown, 'Optical detection of magnetic resonance in a single molecule', Nature 363, 244-245 (1993).