Matrix Isolation Spectroscopy

The aim of this research project is the detection of a single ytterbium (Yb) atom isolated in a solid cryogenic matrix. Over the last 50 years, cryogenic matrices made from solid rare gases (RGs) have been used as hosts to study a wide variety of species difficult to maintain in the gas-phase1. Typically, the techniques used have been those of Matrix Isolation Spectroscopy (MIS) : a large population of the guest species is co-deposited with a rare gas on to a cold substrate to form a solid and then characterized through absorption or emission spectroscopy.

Starting with pentacene (Pc) in 1989, it has been possible to detect exactly one molecule in the solid state2. This is a powerful technique because it removes the ensemble averaging present in standard MIS. The actual values of parameters (e.g. the absorption frequency and lifetimes of the excited states) corresponding to individual molecules are always distributed in the solid state, due to short-range disorder or defects in the matrix. In standard MIS the details of the distribution of these parameters usually disappear upon averaging, whereas single molecule spectroscopy (SMS) gives more information about microscopic structure, allowing the construction of a frequency histogram of the actual distribution of values. Pure quantum effects such as photon bunching3, anti-bunching4, and spectral jumps5,6 have therefore been extensively and more easily studied using SMS.

A natural progression would be to investigate these and other quantum effects using isolated single atoms which are preferable to molecules because they have no intramolecular interactions. To date, select single atoms can be studied through the use of carefully loaded magneto-optical traps7. However, a chemically inert matrix is expected to be an ideal host for single atom experiments, offering longer observation times with only minimal perturbation to the internal structure of the guest atom. In the present case, Yb has been chosen as the test atom because of its simple electronic structure and because it is currently a species of interest for high precision experiments, such as the solid state search for electric dipole moments (EDM) and tests of time reversal symmetry8.

The basic conditions for detection of a single isolated atom through fluorescence spectroscopy are well known: i) only one atom is in resonance in the volume defined by the laser beam; ii) emission from this atom has a signal-to-noise ratio greater than one when averaged over the observation time.

Figure 1. Diagram of the position of the window relative to the cold head . At a temperature of 4 K, all the rare gases will condense and form a matrix suitable for isolating Yb

Our preliminary experiments indicate that the above conditions can be satisfied for Yb isolated in solid rare gas matrix such as Ne. Condition (i) can be achieved through careful preparation of the sample. The matrix will be grown by directing a flux of Ne gas onto a CaF2 window thermally anchored to the cold head of a closed-cycle refrigerator or helium dewar. A flux of Yb atoms will be generated from an effusion oven and directed at the matrix during its growth. The Yb flux can be controlled by temperature to achieve ultralow concentrations (~10-10 M/L) so that when the laser is focused down to ~10 µm3 only a few atoms are present in this volume. As discussed below, condition (ii) can be achieved using high efficiency optics (e.g. a confocal microscope) to collect the fluorescence and a cooled PMT or APD in photon counting mode for detection.

The simple electronic structure of the Yb atom, consisting of spin-singlet and -triplet states, favors single atom detection1. The frequency of the strong 6s2 1S0 → 6s6p 1P1 transition in the solid rare gases (390-405 nm) is easily accessible to powerful UV lasers. Optical pumping of this transition produces fluorescence from the Yb triplet manifold (6s6p 3PJ →6s2 1S0) at very distinct frequencies (> 540 nm) which can easily be separated from scattered laser light using filters.

Figure 2: Emission spectra for Yb isolated in a Ne matrix excited at 390 nm.

Figure 3. Principal optical parts for detection of a single Yb , including laser, SPAD, Emission Filter and Objective lens.

A further important consideration is the size of the Yb absorption cross section, σA, in isolation. The rate of pumping of the 6s2 1S0 → 6s6p 1P1 transition is then given by σAФ, where Ф is the photon flux of the laser. Previous work shows that the absorption crosssection in isolation at 4K is little changed from to its gasphase value2 i.e. σA ~ 3λ2/(2π) = 7.6 x 10-10 cm2. To know the probability of photon emission per absorption event, the fluorescence quantum yield, φF, must be taken into account.

A remarkable feature of the Yb/RG systems is the enhanced high singlet-to-triplet state intersystem crossing, several orders of magnitude greater than In the gas phase (as is confirmed by the complete quenching of the 6s6p 1P1 → 6s2 1S0 emission)9. Assuming a value of φF equal to unity, a laser power of 2 mW/cm2 and a minimal collection efficiency (3 % of the 4π solid angle), would be sufficient to produce a maximum emission count rate of 104 counts/0.1, which is on the order of that for the single atom detection of Cs in the gas phase10.


  1. M. V. Thiel, E. D. Becker, and G. C. Pimental, J. Chem. Phys. 27, 486, 1957.
  2. T. Plakhotnik, E. A. Donley, and U. P. Wild, Annu. Rev. Phys. Chem. 48, 181, 1996.
  3. Basché Th, Moerner W. E., Orrit M., Talon H. ,1992. Phys. Rev. Lett. 69:1516–19
  4. Kummer S, Mais S, Basché Th. 1995. J. Phys. Chem. 99:17078–81
  5. Ambrose W. P., Baché Th, Moerner W.E. J. Chem. Phys. 95:7150–63
  6. Ambrose W. P., Moerner W. E. 1991. Nature 349:225–27
  7. M. Brith and O. Schnepp, J. Chem. Phys., 39, 2714 (1963).
  8. L. A. Andrews and G. C. Pimentel, J. Chem. Phys., 47, 2905 (1967).
  9. C.-Y. Xu et al., Physics Review Letters, 107(9):093001
  10. Z. Hu and J. Kimble, Opt. Lett. 19, 1888, 1994.