Ions shine on - uses of rare earth ions
The range of uses rare earth ions provide for technological development are seemingly endless. Professor Daniel Hewak, head of the Novel Glasses for Optoelectronic Devices Research Group at The University of Southampton, UK shows how this valuable commodity is already changing the world.
Since the invention of the laser, an important and growing application of rare earth ions has been an active medium for light generation. Most rare earth ions provide fluorescence; when excited with a light source they emit at characteristic wavelengths according to their electronic energy levels and the properties of the host material. Among the fifteen rare earth elements, there are hundreds of possible emission bands from UV wavelengths to infrared.
Working the web
While most people are now aware of the internet’s power to transfer information almost instantly around the globe, few people realise that at the heart of the internet are amplifiers, based on the rare earth ion erbium, which boosts the intensity of the signals as they propagate through the world wide web of optical fibres.
Nature has been kind and provided a rare earth ion that can provide this emission precisely at the wavelengths used by telecommunication networks. The emission from the rare earth ion combines with the optical signal and boosts its intensity without the need to reshape, regenerate or electrically amplify the signal. As shown schematically (see figure, below), when a length of optical fibre doped with a few hundred parts per million of the erbium ion is pumped with a laser diode, here operating at 980nm, its energy raises electrons within the shell of the erbium to an excited state. When a weak optical signal at the telecommunications wavelength of 1,550nm interacts with this excited ion, its own signal is boosted in intensity.
The erbium-doped optical fibre amplifier was discovered in the UK at the University of Southampton in 1986. The importance of this device to the telecommunications industry was immediately recognised and efforts in Southampton and Bell Laboratories in New Jersey, USA, quickly lead to the amplifier technology used today.
The host containing the rare earth dopants plays an important role in influencing the characteristics of the light emitted. For example, in a glass host such as an optical fibre there is often no natural place within the disordered random structure of the glass for the rare earth ion to sit and, therefore, concentrations of the dopant are low; long fibres several metres in length are required to effectively amplify a signal. In addition, the disordered glass structure means that each rare earth ion encounters a slightly different local environment. It is this environment; in particular the refractive index and the bond structure surrounding the ion that determines the emission wavelength defined by the electronic structure of the ion. With each ion in glass seeing slightly different surroundings, a rare earth ion emitting light from such a host will have a relatively broad emission spectrum.
Cutting edge lasers
In a crystalline host, things change considerably. A crystalline structure provides long-range order and symmetry and, if the unit cells that fill the lattice structure of the crystal are of a similar size to the rare earth ion there is a natural site for them to sit within the crystal. One of the first solid-state lasers to be developed used a crystal of yttrium aluminium garnet (YAG) doped with neodymium. Several per cent of neodymium could be easily incorporated with the crystal allowing delivery of compact devices much shorter than the long lengths of optical fibre required for amplification. The regular crystalline structure also ensured that each ion encountered a similar environment, resulting in much narrower, spectrally pure emission compared the broader spread of wavelengths seen with neodymium in a glass host.
Neodymium in both crystalline and glass hosts is one of the most important active laser materials. Providing laser output at wavelengths just outside the visible spectrum around 1,000nm, these lasers find applications in material processing, range finding, laser marking, micromachining, surgery and a wide range of research activities. In addition, these highly stable and reliable lasers can be used to pump other lasers or combined with frequency doubling crystals to produce a visible green 532nm beam, as is done in a laser pointer.
Nd-doped glass lasers are readily scaled up and therefore used in extremely high power and high energy applications where terawatt power levels can be achieved through multiple beam devices. These are lasers being pioneered for laser fusion (inertial confinement fusion) experiments underway at the Laboratory for Laser Energetics at the University of Rochester, New York, among others. The laser itself stands 10m tall and is 100m in length and consists of 60 laser beams focused onto a single one millimetre target. Here host effects were critical to achieve such exceptionally high laser powers, new phosphate hosts were developed through collaboration between Lawrence Livermore National Laboratory in California and Hoya Glass Corporation in Japan. The resulting Nd-doped phosphate glass offered a thermal performance, low nonlinearity and high damage threshold thereby ensuring that the laser power did not significantly alter the properties of the host. This glass formulation LHG-8 is now used throughout the world as the preferred host for neodymium in ultra-high power glass lasers.
Like the erbium-doped fibre amplifier, rare earth doped fibres can also form fibre lasers. Today’s optical fibre lasers can generate kilowatts of power from a single hair-thin thread of ytterbium-doped glass and are revolutionising manufacturing, especially high-speed metal cuttings. They are replacing inefficient and bulky CO2 lasers with compact and efficient solid state devices that offer stable beam quality. Optical fibre-related products are not only penetrating existing markets but, more significantly, they expand the application into areas that are impossible by conventional technologies.
Into the Infrared
Laser technology has matured in the visible and near infrared to the extent that their use is ubiquitous. However, beyond about 2,000nm laser technology falters. Conventional host materials are no longer transparent at the lasing wavelength and the fight between radiative and non-radiative emission becomes fierce. The technologically important mid-infrared covers two atmospheric transmission windows one of which is around 3,000–5,000nm. At some bands within this range, lasers easily propagate through air without absorption by water vapour or atmosphere gases. At other wavelengths, the molecular fingerprints of many gases, liquids and solids have their absorption spectra at mid-infrared wavelengths making lasers ideal for sensing and monitoring our air and water, evaluating contaminants and probing the chemical makeup of materials.
Defence, aerospace and industrial applications in lasers are significant within this wavelength range. Medical applications are also important, particularly in ophthalmic and dental use. Lasers at wavelengths around 2.9 microns are strongly absorbed by water whereas at 6.45 microns laser power is easily absorbed by proteins. Use of lasers at these wavelengths provides clean cutting margins and minimal collateral damage.
However, lasers operating in the mid-IR are challenging to realise. Despite the multitude of possible lasing wavelengths many transitions are very low in efficiency and the majority of pump power is lost. The figure below shows the numerous absorption and emission bands for the rare earth praseodymium. The absorption and emission spectra produced by rare-earth ions varies with the host material but the general distribution of the bands still remains essentially the same, allowing decisive identification of a rare earth ion from its optical signature.
Absorption spectra of 1.5mol% praseodymium (Pr3+) doped in gallium lanthanum sulphide Ga:La:S) glass. The inset shows the corresponding emission bands from each energy level. Non-radiative decay routes are not shown
At mid-infrared wavelengths competition between radiative and non-radiative decay becomes important; in addition to laser emission there are now many non-radiating routes for an excited ion to dissipate its energy (see 'Spectroscopy of rare earth ions', below).
As for visible lasers there are many competing laser technologies for the mid-IR. Among compact solid state lasers, transition metals and lead salts offer mid-IR emission while semiconductor lasers can be engineered so that the thickness of the layers alters the characteristic bandgap emission. Many of these devices however require cryogenic cooling to operate although the technology is advancing rapidly. Here rare earths are holding their own and laser hosts based on chalcogenide glasses, glasses which are metallic alloys of sulphur, selenium and/or tellurium, provide both the mid-IR transmission needed and the low phonon energy that ensures efficient radiative decay. Work by Thorsten Schweizer in the 1990s identified 52 viable mid-IR transitions and, within a decade, scientists at the Naval Research Laboratories established the potential of mid-IR lasers based on chalcogenide fibres.
Worldwide research in this field continues apace as scientists work to develop practical rare earth doped lasers and amplifiers across the spectrum. While many of the laser transitions have already been identified, new laser host materials and laser geometries are ensuring the ability of the rare earth ions to shine on.
Spectroscopy of rare earth ions
Rare earth ions absorb light, predominately in the visible region of the spectrum through energy states that originate from electrons in the 4f energy level. The excited electrons then dissipate this energy through a variety of means. This can be through spontaneous emission, through fluorescence at wavelengths defined by the energy distribution of the electrons within the absorption band, or through stimulated emission, the precursor to laser action, in which the wavelength and phase of the emitted photon matches that of the one that stimulated the emission. In addition, there are a variety of non-radiative decay mechanisms, dominated by multi-phonon decay. Here the excited state energy is dissipated as phonons and ultimately as heat within the host. Generally, ionic materials made from light elements such as silica glass, will have high non-radiative decay rates whereas in a covalently bonded material, made from heavy elements weakly bonded together, the probability of non-radiative decay will be much lower. In addition to multiphonon, nearby ions in an excited state can also cooperate, sharing their energy through cross relaxation to leave one of the two in an intermediate energy state.
Thanks to co-authors Ganapathy Senthil Murugan and Vincent Leonard. The authors acknowledge the support of the EPSRC Centre for Innovative Manufacturing in Photonics at the University of Southampton, UK.