
77 Massachusetts Avenue, 13-2114
Cambridge, MA 02139, U.S.A.
|
Prof.
Nuh Gedik: 617.253.3420
Students: 617.253.1829
Fax: 617.253.1847
|
Experimental Techniques
Pump Probe Spectroscopy
Pump probe spectroscopy is the simplest experimental technique
used to study ultrafast electronic dynamics. In this technique, an ultrashort
laser pulse is split into two portions; a stronger beam (pump) is used
to excite the sample generating a non-equilibrium state and a weaker beam
(probe) is used to monitor the pump-induced changes in the optical constants
(such as reflectivity or transmission) of the sample. Measuring the changes
in the optical constants as a function of time delay between the arrival
of pump and probe pulses yields information about the relaxation of electronic
states in the sample. Depending on the specific electronic excitations
being studied, wavelengths of the pump and probe beams can either be the
same (degenerate pump-probe) or different (non-degenerate).
Transient Grating Spectroscopy
Pump probe spectroscopy described above is well
suited for measuring the lifetime of electronic excitations with
femtosecond time resolution. In order to measure propagation of
these excitations in real space, we use transient grating spectroscopy.
In this technique, a pair of femtosecond pulses is interfered on
the sample to generate a sinusoidal intensity modulation, that in
turn induces a density grating of photoexcitations. Because the
index of refraction depends on the local excitation density, a periodic
modulation of the index of refraction is formed. The period of this
pattern in real space can be changed either by changing the wavelength
of the laser or the angle between the two beams. An incident probe
pulse on this pattern is therefore both reflected and diffracted.
Measuring the time evolution of both the reflected and diffracted
waves enables us to track the propagation of these excitations in
real space.
Ultrafast Electron Diffraction
Direct determination of structural dynamics requires the ability
of measuring atomic motions with angstrom scale spatial resolution.
Conventional ultrafast optical spectroscopy based on measuring transient
changes in optical constants is sensitive to dynamics of electronic
excitations but can provide only indirect information about structural
dynamics. The spatial resolution in these techniques is also limited
to micron scales due to diffraction limit.
Ultrafast electron diffraction (UED) can directly couple
to structural dynamics and provide sub-angstrom spatial resolution together
with sub-picosecond temporal resolution. The principle of UED is similar
to pump probe spectroscopy. An ultrafast laser pulse is split into two;
the first part of the laser pulse is directly focused on to the sample
to create a non-equilibrium state. To probe the induced structural change,
the second part is frequency tripled and focused on to a photocathode
generating an ultrafast electron packet via photoelectric effect. These
electrons are then accelerated through a high voltage (typically through
30 keV, de Broglie wavelength = 0.07 Å) and diffracted from the
sample.
The relative arrival time of the probing electron packet
and the initiating laser pulse at the sample can be changed by changing
the relative optical path-lengths of the two laser beams. Recording the
diffraction pattern of the electron packet as a function of this time
delay provides both the equilibrium structure and a movie of the structural
evolution with sub-Angstrom spatial resolution (reaching ~0.001 Å level)
and sub-picosecond temporal resolution.
Materials
Quantum Materials
In conventional materials, kinetic energy of the electrons is
much higher than the interaction energy between them. Therefore, simple theories
ignoring electron-electron interactions or treating it as a small perturbation can be
used to describe the properties of these systems very accurately. Materials in which the
interaction energy between the electrons is much higher than the electron kinetic energy
are known as quantum materials or strongly correlated electron systems. These materials
exhibit fascinating properties such as high temperature superconductivity or collosomagnetoresistance
which can not be explained by conventional theories using a perturbative approach to treat interactions.
Cuprate superconductors are classical examples of quantum materials. The undoped parent
cuprates have one electron per copper site, so the band is half filled. The band theory
would predict them to be metals; yet they are anti ferromagnetic Mott insulators! The
reason is that the energy cost of doubly occupying a copper site (U ~ 10 eV) due to strong
coulomb repulsions is much bigger than the kinetic energy gained by hopping (t ~ 1 eV). So
the electrons antiferromagnetically align themselves and localize at copper sites. Cuprates
have several interesting features. Firstly, different degrees of freedoms (i.e. charge, spin and lattice)
are strongly coupled in these systems and interplay between them results in many of their fascinating properties.
Secondly, when these materials are doped chemically they display a complex phase diagram and can have multiple
competing phases. Understanding the nature of coupling between different degrees of freedoms and dynamics of phase
transitions between different competing phases are crucial in understanding the physics behind these materials.
We use the experimental techniques described above to study quantum materials. These novel time-resolved
techniques can provide information that is not accessible to conventional techniques.
|