INSIGHTS | PERSPECTIVES
138 11 JULY 2014 • VOL 345 ISSUE 6193 sciencemag.org SCIENCE
chain crystallites. The T-ULEED experiment revealed that ultrafast melting of the
PMMA follows a hierarchy of events. The
graphene substrate absorbing the laser
pump energy increased its temperature to
540 K within the pump-pulse duration of
80 fs. Energy transfer to the PMMA bilayer
led to an elevated temperature equilibration
with a characteristic time of 43 ps. When
the magnitude of transferred energy (heat)
was sufficient, a loss of PMMA crystallinity,
a change in PMMA chain spacing, and structural relaxation to the amorphous phase occurred on characteristic times of hundreds
The innovative part of the T-ULEED experiment is the short-pulse illumination of
a tungsten nanotip to produce the electron
pulses. In recent years, it has been shown
that few-cycle laser pulses focused to the
apex of metal tips can induce field-driven
photoemission of ultrashort electron pulses
(3, 4). This phenomenon has now been explored in cases ranging from multiphoton
ionization to above threshold ionization and
strong-field tunneling emission, including
carrier-envelope phase effects (5).
Exploiting strong-field emission from a
metal nanotip opens up the new frontier of
ultrafast LEED techniques by substantially
miniaturizing the electron diffraction apparatus compared with the high-energy diffraction methods that have recently come
to fruition (6–8). Ultrafast electron or x-ray
diffraction (9) techniques have provided insight into the structural dynamical events
underlying processes as diverse as photoin-duced chemical reactions in the gas phase,
charge transfer in bulk systems, and melting
of metal or semiconductor superstructures.
The T-ULEED method of Gulde et al.
widens the field of ultrafast structural dynamics to surface science. Previous ultrafast electron diffraction methods relied
on sophisticated experimental set-ups to
produce the intense, short electron pulses
needed to probe irreversible structural
changes or to achieve femtosecond time
resolution. Achieving similar time resolution with the T-ULEED experiment will be
the next challenge. ■
1. M.Guldeetal.,Science 345, 200(2014).
2. C. Davisson, L. H. Germer, Phys.Rev. 30, 705 (1927).
3. P. Hommelhoff, C. Kealhofer, M. A. Kasevich, Phys.Rev.
Lett. 97, 247402 (2006).
4. C. Ropers, D. R. Solli, C. P. Schulz, C. Lienau, T. Elsaesser,
Phys. Rev. Lett. 98, 043907 (2007).
5. B. Piglosiewicz et al ., Nat. Photonics 8, 37 (2014).
6. A.H.Zewail,Science 328, 187(2010).
7. R. J. D. Miller, Annu. Rev. Phys. Chem. 65, 583 (2014).
8. F. O. Kirchner, A. Gliserin, F. Krausz, P. Baum,Nat.
Photonics 8, 52 (2014).
9. T.Elsaesser,M. Woerner, J. Chem. Phys. 140,020901
The discovery of superconductiv- ity in UPt3 (1, 2) immediately led to ideas and speculations about how the electrons pair up and condense into a superconducting state. Spe- cifically, questions have surrounded
the symmetry of the order parameter in
these materials (a measure of the density of
the condensed electron pairs, and of their
phase). From the outset, it was suspected
that these so-called heavy-fermion systems
were “odd-parity” or “p-wave” superconductors (3); that is, instead of the electrons pairing up into spin-singlet states with opposite
spins as in conventional superconductors,
they would form spin-triplet states where
the spins of the paired electrons point in
the same direction. To date, consensus concerning the microscopic pairing mechanism
applicable to these heavy-fermion systems
has been lacking. On page 190 of this issue,
Schemm et al. ( 4) report results of magneto-optical experiments that define the nature
of the superconducting pairing in these
compounds. The Kerr-rotation experiments
provide compelling evidence in favor of the
odd-parity mechanism, and show that UPt3
belongs to the same universality class as superfluid 3He and the ruthenates.
Various mechanisms favoring odd-parity
superconductivity have been presented,
either mediated by ferromagnetic spin-fluctuations (5) or by Hund’s rule coupling
(6–9). Other models have favored singlet
superconductivity (10). Historically, the observation of strong mass renormalization in
UPt3 and other heavy-fermion metals suggested a close analogy with 3He, which is
also a Fermi liquid characterized by strong
mass renormalization. Taken together with
the high-spin susceptibility, the possibility
was explored that the pairing symmetry in
UPt3 would also have similarities with 3He,
which is of the triplet variety.
Before the unique playground offered
by these unconventional superconductors
can be exploited, several questions must be
answered. One concerns the interactions
responsible for selecting the p-wave (trip-
let) pairing with a pair spin, S = 1, rather
than the more conventional s-wave (sin-
glet) pairing in which the pairs have zero
spin. It could be that p-wave pairing occurs
much more often than was generally as-
sumed, but has gone unnoticed because the
experiments were unable to detect the p-
wave pairing aspects. In this regard, one of
the first indicators of unconventional pair-
ing was the extreme sensitivity to impuri-
ties. So what other tests, if any, of p-wave
superconductivity are possible beyond the
existing ones, including the extremely high-
sensitivity magneto-optical measurements
that Schemm et al. report?
It is also desirable to distinguish between
the various proposed mechanisms for the
pairing. Recently, Hund’s rule coupling has
become popular in the description of correlated electron systems such as the iron-based pnictide superconductors as so-called
Hund’s rule metals. In Hund’s metals, the
mass of the electrons is enhanced because
of the Hund’s rule interaction, J, that tends
to align the spin of electrons when they occupy the same uranium atom.
Another question is whether the S = 1
character of the paired electrons can be exploited in potential spintronics applications
that would use the flow of a supercurrent
carried by condensed pairs and carrying
with it a magnetic moment.
Most superconducting materials are of
the conventional S = 0 variety in which
1Département de Physique de la Matière Condensée, Ecole
de Physique, Université de Genève, Quai Ernest-Ansermet
24, Genève, Genève CH-1211, Switzerland. 2Department of
Physics and Astronomy, University of British Columbia, 6224
Agricultural Road, Vancouver, British Columbia V6T 1Z1,
Canada. E-mail: email@example.com; sawatzky@
An optical twist for triplet
“… finding the correct
balance between these
interaction channels may be
sufficient to form an exotic
a desired set of properties.”
By Dirk van der Marel1 and
George Albert Sawatzky2
Optical measurements may help reveal the secrets of exotic
superconductivity and manipulate the pair condensate