, 463 (2011);
, et al.Zubin Jacob
Plasmonics Goes Quantum
This copy is for your personal, non-commercial use only.
clicking here.colleagues, clients, or customers by
, you can order high-quality copies for your
If you wish to distribute this article to others
The following resources related to this article are available online at
here.following the guidelines
can be obtained by
Permission to republish or repurpose articles or portions of articles
Updated information and services,
): December 21, 2011 www.sciencemag.org (this infomation is current as of
version of this article at:
including high-resolution figures, can be found in the online
, 2 of which can be accessed free:
cites 14 articles
This article appears in the following
registered trademark of AAAS.
is a Science2011 by the American Association for the Advancement of Science; all rights reserved. The title
CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the Science
on December 21, 2011
www.sciencemag.org SCIENCE VOL 334 28 OCTOBER 2011
opportunities for exploitation in the devel-
opment of novel heparin-based therapeutics
( 10– 12). Synthetic chemistry and chemoen-
zymatic approaches represent interrelated
pathways that could deliver such compounds
for clinical evaluation. Modular approaches
to the full chemical synthesis of small librar-
ies of tetra- and hexasaccharides have been
described ( 13– 15), and increasingly larger
individual target structures are being synthe-
sized ( 16, 17).
The key questions now for LMW hepa-
rin production as anticoagulants are whether
the new chemoenzymatic processes can be
scaled up to meet industrial production levels
(an estimated 10 to 20 tonnes per year world-
wide) and can compete on cost with the natu-
ral product route. Although the chemoenzy-
matic approach can in principle be extended
to make larger heparins and more structur-
ally diversifi ed “heparan sulfate–type” com-
pounds, the heparin saccharides generated by
Xu et al. took advantage of natural specifi ci-
ties of some of the enzymes, which imposes
some limitations on product range. Careful
design of the sequence of modifi cations, and
perhaps additional chemical control strate-
gies, will likely be needed to permit gener-
ation of some of the structural variants that
may be required. Nevertheless, the prospects
for fully defi ned heparin and related com-
pounds seem very bright, with huge potential
to secure safe production routes for replace-
ment of animal-derived heparins, as well as
providing a novel class of compounds with
important therapeutic applications.
1. T. K. Kishimoto et al., N. Engl. J. Med. 358, 2457 (2008).
2. U.S. Food and Drug Administration, Information on Adverse
Event Reports and Heparin, www.fda.gov/DrugSafety/
4. H. Liu et al., Nat. Prod. Rep. 26, 313 (2009).
5. Y. Xu et al., Science 334, 498 (2011).
6. M. Guerrini et al., Nat. Biotechnol. 26, 669 (2008).
7. M. Guerrini et al., Proc. Natl. Acad. Sci. U.S.A. 106,
8. M. Petitou, C. A. A. van Boeckel, Angew. Chem. Int. Ed.
43, 3118 (2004).
9. J. R. Bishop, M. Schukszand, J. D. Esko, Nature 446,
10. R. Lever, C. P. Page, Nat. Rev. Drug Discov. 1, 140
11. J. L. Stevenson et al., Clin. Cancer Res. 11, 7003 (2005).
12. S. J. Patey et al., J. Med. Chem. 49, 6129 (2006).
13. J. L. de Paz et al., J. Am. Chem. Soc. 128, 2766 (2006).
14. S. Arungundram et al., J. Am. Chem. Soc. 131, 17394
15. Z. Wang et al., Chemistry 26, 8365 (2010).
16. C. L. Cole et al., PLoS ONE 5, e11644 (2010).
17. Y.-P. Hu et al., Nat. Chem. 3, 557 (2011).
Plasmonics Goes Quantum
Zubin Jacob 1 and Vladimir M. Shalaev 2
A seamless interface between the two can
guarantee the use of light to overcome issues
related to the resistive time delay of electrons
within integrated circuits. However, a funda-
mental incompatibility arises between photo-
nics and nanometer-scale electronics because
light breaks free when confi ned to sizes below
its wavelength. Instead, coupling light to the
free electrons of metals can lead to a quasipar-
ticle called a plasmon, with nanometer-scale
mode volumes. The resulting possibility of
effi ciently interfacing photonics and
nanoelectronics has been the impetus
for the fi eld of plasmonics ( 1). Recent
work has shown that these nanoscale
plasmons, which can transmit classical
information with unprecedented band-
width, are also naturally conducive to
quantum information processing ( 2).
Tiny light emitters like quantum
dots and molecules interact excep-
tionally well with low–mode-volume
A combined plasmonics and metamaterials
approach may allow light-matter interaction
to be controlled at the single-photon level.
ight in a silica fi ber and electrons in sil-
icon are the backbones of current com-
munication and computation systems.
plasmons, but not with the photonic modes
of a conventional optical fi ber ( 2). Although
a host of nanoscale metallic particles can
exhibit localized plasmonic behavior at a
particular resonant excitation frequency, the
nanowire supports a single propagating plas-
mon mode for a wide range of frequencies.
It is similar to a single-mode optical fi ber,
but with the advantage that the mode is con-
fi ned to the nanoscale dimensions of the wire.
This leads to strong coupling with isolated
broadband emitters at room temperature. An
excited quantum dot near a metallic nanowire
will almost always spontaneously emit a sin-
gle photon into this fundamental plasmonic
mode. Such robustness is central to reli-
able technological exploitation of quantum-
mechanical rules that are otherwise governed
by probabilities (see the fi gure, panel A).
The quantum properties of this single
plasmonic oscillation were demonstrated
with antibunching statistics ( 2) as well as
wave-particle duality ( 3). This is not surpris-
ing because plasmons have been shown to
preserve quantum information in the light
used to generate them. Experiments have
conclusively proven that conventional plas-
mons on a nanohole array or a gold metal
strip can show exotic quantum
properties such as entanglement
( 4) and squeezing ( 5).
Remarkably, despite the deco-
herence expected due to collisions,
the millions of electrons making up
the plasmon conspire to carry the
quantum bit originally encoded in
the photon ( 6). Furthermore, this
nonclassical information survives
conversion and can be faithfully
recovered ( 5). There is little doubt
that the plasmon is uniquely poised
to play a role in future nanoscale
quantum information processing.
A related question is about
what happens in the limit of a
1Department of Electrical and Computer Engi-
neering, University of Alberta, Edmonton, Alberta
T5K2M3, Canada. 2Department of Electrical
and Computer Engineering, Birck Nanotechnol-
ogy Center, Purdue University, West Lafayette,
IN 47906, USA. E-mail: email@example.com,
Make it quantum. Building blocks of an integrated nanoscale quantum
information system. (A) The nanowire supports a single plasmonic oscil-
lation conceptually similar to a single-mode optical fi ber. However, the
nanoscale mode volumes of the plasmon lead to strong coupling with the
quantum emitter. (B) An unorthodox approach of enhancing light-matter
interaction is by tailoring the dielectric constant of a medium so that it is
dielectric in one direction and metallic in another. The resulting hyperbo-
loidal dispersion relation supports infi nitely many electromagnetic states
for channeling light into a single-photon resonance cone.
Published by AAAS
on December 21, 2011
28 OCTOBER 2011 VOL 334 SCIENCE www.sciencemag.org
How Proteins Fold
Tobin R. Sosnick 1 and James R. Hinshaw 2
of this issue, Lindorff-Larsen et al. ( 1) use
state-of-the-art molecular dynamics (MD)
simulations to elucidate the folding mech-
anisms of 12 different proteins. On page
512 of this issue, Stigler et al. ( 2) study the
folding and unfolding of single calmodulin
domains with single-molecule force spec-
troscopy. The results provide remarkable
views of the folding process and address
basic questions, such as whether proteins
fold along pathways.
Computational and experimental results provide
support for defi ned protein folding pathways.
large ensemble of emitters near a metallic
nanoparticle. The preferential emission into
a single plasmonic mode also makes pos-
sible the concept of a spaser, an amplifi er
and coherent generator of plasmons ( 7). To
understand plasmons in complex nanoparti-
cle architectures for spasers or other appli-
cations such as single-molecule detection,
it is important to incorporate a quantum-
mechanical model of the electron density
in the analysis. Effects such as tunneling of
electrons between coupled nanostructures
can considerably affect the nature of plas-
mons ( 8). Such a microscopic approach will
be critical to develop a complete understand-
ing of quantum plasmonics.
Efforts in quantum optics have been
directed toward overcoming decoherence
effects and achieving scalability of quan-
tum bits (qubits) for practical applications.
One approach is to achieve parallelism and
communication between quantum bits of dif-
ferent nature (e.g., spin qubits and photonic
qubits; akin to our current use of optoelec-
tronics for computation and communica-
tion). The nitrogen-vacancy (NV) center in
diamond is a promising choice for the robust
solid-state quantum bit because it can show
single-photon emission as well as long spin
coherence times ( 9). The ability to make
these two degrees of freedom interact rests
on effi cient single-photon emission beyond
that available in bulk diamond NV centers.
Resonant cavity approaches to enhancing the
optical emission are incompatible with these
sources, which have a broad emission spec-
trum. The broadband enhancement of spon-
taneous emission enabled by nanoplasmonic
approaches allows the possibility of coupling
to such emitters, which was otherwise diffi -
cult to achieve by conventional quantum opti-
cal techniques ( 10).
Another unorthodox approach of enhanc-
ing the nanoscale light-matter interaction in
a broad bandwidth is to provide the quantum
emitter with a plethora of electromagnetic
states ( 11). Current nanofabrication technol-
ogies allow the engineering of the dielectric
constant with metamaterials, transforming
the space perceived by light to be metallic in
one direction and dielectric in another. This
lifts the restriction on the well-known closed
spherical dispersion relation of an isotropic
medium into a hyperboloid, leading to elec-
tromagnetic states unique to the metamate-
rial ( 12, 13). An infi nite number of metama-
terial states can lie on this hyperboloid (in the
low-loss, effective-medium limit), increas-
ing the interaction with the quantum emit-
ter while simultaneously channeling the light
into a subdiffraction single-photon resonance
cone ( 12) (see the fi gure, panel B). Currently,
losses present a formidable challenge to prac-
tical applications, but the new class of alter-
nate plasmonic materials can lead to quan-
tum-vacuum engineered devices with these
“hyperbolic” metamaterials ( 14).
The future of nanophotonics is bright,
with many possibilities of interfacing with
quantum optics to address challenges of qubit
scalability and communication. One topic to
be addressed in the near future is single-pho-
ton switching and routing. Single photons do
not talk to each other, but efforts are under
way to use plasmon-mediated interactions
for this purpose ( 15). It is quite likely that
the hybrid excitation that combines photons
and electrons will be the carrier of choice in
future quantum information systems.
References and Notes
1. M. L. Brongersma, V. M. Shalaev, Science 328, 440 (2010).
2. A. V. Akimov et al., Nature 450, 402 (2007).
3. R. Kolesov et al., Nat. Phys. 5, 470 (2009).
4. S. Fasel et al., Phys. Rev. Lett. 94, 110501 (2005).
5. A. Huck et al., Phys. Rev. Lett. 102, 246802 (2009).
6. E. Altewischer et al., Nature 418, 304 (2002).
7. D. J. Bergman, M. I. Stockman, Phys. Rev. Lett. 90, 027402
8. J. Zuloaga et al., Nano Lett. 9, 887 (2009).
9. I. Aharonovich et al., Nat. Photonics 5, 397 (2011).
10. A. Huck et al., Phys. Rev. Lett. 106, 096801 (2011).
11. Z. Jacob et al., http://arxiv.org/abs/0910.3981 (2009).
12. Z. Jacob et al., Appl. Phys. B Lasers Opt. 100, 215 (2010).
13. M. A. Noginov et al., Opt. Lett. 35, 1863 (2010).
14. A. Boltasseva, H. A. Atwater, Science 331, 290 (2011).
15. D. E. Chang et al., Nat. Phys. 3, 807 (2007).
16. This work was supported in part by Army Research Offi ce–
Multidisciplinary University Research Initiative (ARO-MURI)
grant W911NF-0910539, ONR-MURI grant N00014-
010942, and Natural Sciences and Engineering Research
Council of Canada Discovery grant 402792. Z.J. acknowl-
edges input from W. Newman.
wo reports in this issue probe sin-
gle protein chains as they spontane-
ously unfold and refold. On page 517
The Shaw group previously succeeded
in modeling the folding of a 35-residue pro-
tein in the presence of water molecules ( 3).
Lindorff-Larsen et al. now show that these
methods are suitable for probing the folding
of larger, more complex proteins. In addi-
tion to matching experimental folding rates,
the authors obtain native-like models for 12
proteins, which contain helices and sheets
and are up to 80 residues long. For most of
the proteins, the trajectories contain discrete
transitions between the native and unfolded
states. This behavior is consistent with bar-
rier-limited cooperative folding, the hallmark
of the experimental folding reaction.
Whether folding occurs along a diverse
set of routes elicits diverse opinions, with
many researchers favoring extensive path-
way heterogeneity due to the complexity of
the system ( 4). Yet, Lindorff-Larsen et al.
fi nd that for nine of their proteins, hetero-
geneity is minimal, with the routes typically
sharing over 60% of their native contacts.
They conclude that these routes are best
viewed as variations of a single folding path-
way. This lack of pathway diversity is consis-
tent with experimental studies where transi-
tion state heterogeneity was not observed ( 5).
However, in the simulations of two β sheet–
containing proteins, the order of strand for-
mation can vary. For a protein G variant, the
observation of two pathways is consistent
with experimental work, which found that
the folding order of the two hairpins can be
manipulated ( 6). Generally, symmetric ( 7, 8)
and multidomain proteins are strong candi-
dates for multiple pathways, because differ-
ent portions can form fi rst.
1Department of Biochemistry and Molecular Biology, Insti-
tute for Biophysical Dynamics, Computation Institute, Chi-
cago, IL 60637, USA. 2Department of Chemistry, University
of Chicago, Chicago, IL 60637, USA.
Published by AAAS
on December 21, 2011