Metallic and insulating oxide interfaces controlled by electronic correlations.

Abstract

The formation of two-dimensional electron gases (2DEGs) at complex oxide interfaces is directly influenced by the oxide electronic properties. We investigated how local electron correlations control the 2DEG by inserting a single atomic layer of a rare-earth oxide (RO) [(R is lanthanum (La), praseodymium (Pr), neodymium (Nd), samarium (Sm), or yttrium (Y)] into an epitaxial strontium titanate oxide (SrTiO(3)) matrix using pulsed-laser deposition with atomic layer control. We find that structures with La, Pr, and Nd ions result in conducting 2DEGs at the inserted layer, whereas the structures with Sm or Y ions are insulating. Our local spectroscopic and theoretical results indicate that the interfacial conductivity is dependent on electronic correlations that decay spatially into the SrTiO(3) matrix. Such correlation effects can lead to new functionalities in designed heterostructures.

Full text

Page 1

37. M. Halic et al., Science 312, 745 (2006).
38. K. Mitra et al., Nature 438, 318 (2005).
39. J. R. Jagath, M. V. Rodnina, W. Wintermeyer, J. Mol. Biol.
295, 745 (2000).
40. B. Weiche et al., J. Mol. Biol. 377, 761 (2008).
41. Single-letter abbreviations for the amino acid residues
are as follows: A, Ala; G, Gly; Q, Gln; and R, Arg.
42. We thank K. Zhou for excellent technical assistance
and help with crystal preparation during the early stages
of the project. Initial crystallographic analysis was
performed at beamline 8.2.2 at the Advanced Light
Source (ALS), Lawrence Berkeley National Laboratory;
we acknowledge C. Ralston for outstanding technical
assistance at the ALS. Crystallographic data were
collected at the beamline X06SA at the Swiss Light Source
(SLS). We thank A. Brunger for the prerelease version
of CNS and for helpful comments on the refinement,
C. Schulze-Briese and T. Tomizaki for their outstanding
support at the SLS, T. Maier and S. Klinge for critical
discussion and reading of the manuscript, and T. Maier
and M. Leibundgut for help and assistance with data
collection and solving the structure. S.F.A. was funded
initially by the Howard Hughes Medical Institute and
currently by an ETH postdoctoral fellowship, N.S. is
funded by Boehringer Ingelheim Fonds, and K.S. is
funded by NIH grant GM078024 to S.S. This work was
supported in part by the Howard Hughes Medical Institute
(J.A.D.) and by the Swiss National Science Foundation
(SNSF) and the National Center of Excellence in Research
(NCCR) Structural Biology program of the SNSF. Atomic
coordinates and structure factors for the SRP:SR crystal
structure have been deposited with the Protein Data Bank
under accession code 2xxa.
Supporting Online Material
www.sciencemag.org/cgi/content/full/331/6019/881/DC1
Materials and Methods
Figs. S1 to S8
Tables S1 and S2
References
13 August 2010; accepted 18 January 2011
10.1126/science.1196473
REPORTS
Metallic and Insulating Oxide
Interfaces Controlled by
Electronic Correlations
H. W. Jang,1D. A. Felker,2C. W. Bark,1Y. Wang,3M. K. Niranjan,3C. T. Nelson,4Y. Zhang,4,5
D. Su,6C. M. Folkman,1S. H. Baek,1S. Lee,1K. Janicka,3Y. Zhu,6X. Q. Pan,4D. D. Fong,7
E. Y. Tsymbal,3M. S. Rzchowski,2C. B. Eom1*
The formation of two-dimensional electron gases (2DEGs) at complex oxide interfaces is directly
influenced by the oxide electronic properties. We investigated how local electron correlations
control the 2DEG by inserting a single atomic layer of a rare-earth oxide (RO) [(R is lanthanum
(La), praseodymium (Pr), neodymium (Nd), samarium (Sm), or yttrium (Y)] into an epitaxial
strontium titanate oxide (SrTiO3) matrix using pulsed-laser deposition with atomic layer control.
We find that structures with La, Pr, and Nd ions result in conducting 2DEGs at the inserted
layer, whereas the structures with Sm or Y ions are insulating. Our local spectroscopic and
theoretical results indicate that the interfacial conductivity is dependent on electronic
correlations that decay spatially into the SrTiO3matrix. Such correlation effects can lead to
new functionalities in designed heterostructures.
A
multilayers (1), superlattices (2–4), and ultrathin
films(5,6).Intheseartificialstructures,the inter-
faces play a prominent role in determining the
functionalities of the structures and their applica-
tions (7). A recent example is the discovery of
two-dimensional electron gases (2DEGs) at the
interface between complex insulating oxides (8)
such as LaAlO3/SrTiO3(9, 10), LaTiO3/SrTiO3
(2), and LaVO3/SrTiO3(11) heterostructures, in
dvanced deposition techniques enable the
growth of epitaxial heterostructures with
atomically controlled interfaces such as
which the 2DEG is confined near the LaO/TiO2
interface. Magnetic and superconducting ground
states of the 2DEG have been identified (12–14),
and applications to field-effect transistors and tun-
nel junctions have been demonstrated (15–17).
Theoretical work on LaTiO3/SrTiO3super-
lattices (18) suggests that for a several-unit-cell-
thick LaTiO3layer, the LaTiO3/SrTiO3interface
regionismetallic;however,nonmetallicbehavior
dominates in the LaTiO3region away from the
interface, resulting from strong electron correla-
tions similar to those found in bulk LaTiO3. In
other bulk rare-earth titanates, the effect of elec-
tron correlations depends critically on the rare-
earth ion (19). We used the unique electronic char-
acter ofoxide interfaces,andatomic levelcontrol
oftheir structure andcomposition,todeliberately
manipulate the 2DEG electronic properties.
We studied the effect of strong electron cor-
relations on an oxide 2DEG by inserting a single
atomic layer of RO (R is La, Pr, Nd, Sm, or Y)
intoanepitaxialSrTiO3matrixusingpulsed-laser
deposition with atomic layer control. The RO
layer donates electrons to the conduction band of
SrTiO3. These electrons remain near the inserted
ROlayerdue to Coulomb attraction.Wefindthat
the transport properties of these electrons range
from metallic to insulating, depending critically
1Department of Materials Science and Engineering, University
ofWisconsin–Madison,Madison,WI53706,USA.2Department
of Physics, University of Wisconsin–Madison, Madison, WI
53706,USA.3DepartmentofPhysicsandAstronomy,Nebraska
CenterforMaterialsandNanoscience,UniversityofNebraska–
Lincoln, Lincoln, NE 68588, USA.4Department of Materials
Science and Engineering, University of Michigan–Ann Arbor,
AnnArbor,MI48109,USA.5NationalLaboratoryofSolidState
Microstructures and Department of Materials Science and
Engineering, Nanjing University, Nanjing 210093, P.R. China.
6Center for Functional Nanomaterials, Brookhaven National
Laboratory, Upton, NY 11973, USA.7Materials Science Divi-
sion, Argonne National Laboratory, Argonne, IL 60439, USA.
*To whom correspondence should be addressed. E-mail:
eom@engr.wisc.edu.
Fig. 1. (A) Schematic
representation of a SrTiO3/
1-ML RO/SrTiO3 hetero-
structure. The atomic struc-
ture near the interface is
enlarged. The +1 valent
RO layer donates elec-
trons to neighboring TiO2
planes, leading to the
larger electron density ne
neartheinterface.(B)Typ-
ical RHEED oscillations
for the growth of 1-ML
LaO and 10-uc SrTiO3lay-
ers in sequence on a TiO2-
terminatedSrTiO3substrate.
(C) AFM image of a 10-uc
SrTiO3/1-ML LaO/SrTiO3
heterostructure showing an
atomically smooth surface.
Published in SCIENCE, vol 331, issue 6019 (February 18, 2011): 886-889. DOI: 10.1126/science.1198781
This article is a U.S. government work, and is not subject to copyright in the United States.

Page 2

on the rare-earth ion, and that this dependence
arises from strong electronic correlations.
We grew epitaxial SrTiO3 heterostructures
containing a symmetric TiO2/RO/TiO2interface
(Fig. 1A), resulting in RTiO3-like structure at
the interface. Using pulsed-laser deposition,
the heterostructures were fabricated by depos-
iting either a RO monolayer or a RTiO3unit cell
on a TiO2-terminated SrTiO3substrate, followed
by deposition of a SrTiO3overlayer of varying
thickness (20). A thick SrTiO3overlayer approx-
imates a single RO monolayer embedded in an
infinite SrTiO3matrix. Thicknesses of inserted
1-monolayer (ML)–thick RO and 1-unit-cell (uc)–
thick RTiO3layers were accurately controlled
by monitoring in situ reflection high-energy elec-
tron diffraction (RHEED) intensity oscillations.
Typical RHEED oscillations for the growth of
a 10-uc SrTiO3/1-ML LaO heterostructure on a
SrTiO3substrate are shown in Fig. 1B. The atomic
force microscopy (AFM) image of the surface
of a complete heterostructure (Fig. 1C) shows the
steps and terraces of the original substrate sur-
face, indicating high-quality growth. Microstruc-
ture and electrical properties of both SrTiO3/1-ML
RO/SrTiO3and SrTiO3/1-uc RTiO3/SrTiO3hetero-
structures were almost identical (20). Here, we
focus on the SrTiO3/RO/SrTiO3heterostructures.
We first characterized the dependence of
electrical properties on growth conditions, using
the LaO-based heterostructure, and established
thegrowthconditionsofoxygenpressure(PO2¼
10−3mbar) and temperature (Tgrowth= 550°C) as
optimal (20). These growth conditions were used
to fabricate SrTiO3heterostructures with single
inserted atomic layers of LaO, PrO, NdO, SmO,
and YO. Fig. 2A shows the mobile sheet carrier
concentration nsfor the five different RO layers
asafunctionoftheSrTiO3overlayerthickness.It
is seen that LaO-, PrO-, and NdO-based hetero-
structures become conducting above the critical
thickness of SrTiO3of three or four unit cells.
However, SmO- and YO-based heterostructures
are insulating, even with a SrTiO3 overlayer
thickness of 100 uc. This is summarized in Fig.
2B, which shows the mobile sheet carrier
concentration at fixed overlayer thickness as the
rare-earth ion progresses from La to Y. The nom-
inal room-temperature concentration of mobile
carriers in crossover NdO-based heterostructures
decreasesdramaticallyatlowertemperatures(fig.
S3D), in contrast to the relatively temperature-
independent behavior of the conducting LaO-
based and PrO-based heterostructures. This trend
is analogous to that in bulk RTiO3, where the
effects of electron correlations increase as R is
varied from La to Y (21). The mobilities of all
conducting heterostructures are roughly indepen-
dent of the rare-earth ion, showing a crossover
fromtemperature-dependentphononscatteringat
high temperature to a temperature-independent
value at low temperatures.
Our transport measurements are sensitive to
mobile carriers near the interface. We also inves-
tigated charge transfer from the RO layer to
nearby Ti states with electron energy-loss spec-
troscopy (EELS), sensitive to both mobile and
nonmobilecarriers(2,10).Fora conductingLaO-
based heterostructure, the spatial dependence of
EELS spectra of Ti-L2,3and O-K edges is shown
in Fig. 3B. The EELS spectra are spatially sep-
arated by 0.28 nm, in a line scan across the
interface of a 10-uc SrTiO3/1-ML LaO/SrTiO3
heterostructure (Fig. 3A). Four clear peaks in the
Ti L2,3edge become broader at the interface, with
peak separations less pronounced. We attribute
this broadening to the presence of a Ti3+com-
ponent. Compared with previous reports (2, 10),
therelativelysmallmodulationoftheEELSsignal
at the interface may be related to the low ns
determined from the Hall effect. Our depth profil-
Fig. 2. Dependence of sheet carrier concentration nson the R ion in SrTiO3/1-ML RO/SrTiO3hetero-
structures and the SrTiO3overlayer thickness d. Sheet carrier concentration is plotted as a function of (A)
the thickness of the SrTiO3 overlayer and (B) the RO doping layer for SrTiO3/1-ML RO/SrTiO3
heterostructures. SmO-based and YO-based heterostructures never become conducting, even with very
thick SrTiO3overlayers.
Fig. 3. STEMandEELSanalysis.(A)High-angleannulardarkfield(HAADF)imageofa10-ucSrTiO3/1-ML
LaOfilmgrownonSrTiO3.TherectangularboxrepresentstheregionofEELSlinescans.(B)EELSspectraof
T-L2,3andO-Kedgesobtainedfrom2Dlinescansacrosstheinterfaceshownin(A).Thespacingalongthe
line scan between consecutive EELS spectra is 2.8 Å. The spectra at the LaO layer are highlighted by thicker
lines. For the spectra for Ti L2and L3edges, peak broadening and less pronounced peak splitting at the
interface are clearly observed. (C) HAADF images of 10-uc SrTiO3/1-ML LaO/SrTiO3and 10-uc SrTiO3/1-ML
SmO/SrTiO3heterostructures. Both samples show no obvious defects or dislocations, indicating coherent
interfaces.(D)SelectedareaTi-L2,3EELSspectraobtainedattheinterfacesfor10-ucSrTiO3/1-MLLaO/SrTiO3
and 10-uc SrTiO3/1-ML SmO/SrTiO3heterostructures. The arrow is a guide for comparison.
www.sciencemag.org
SCIENCE
VOL 33118 FEBRUARY 2011
887
REPORTS

Page 3

ing of the Ti3+to Ti4+ratios indicates that the
carriers are confined to within ~1 nm of the inter-
face (fig. S6), in good agreement with recent
theoretical calculations (22).
Fig. 3, C and D, show scanning transmission
electron microscope (STEM) images and selected
area Ti-L2,3EELS spectra, at TiO2planes adjacent
to the interface, for LaO (conducting) and SmO
(insulating) heterostructures. For both heterostruc-
tures, the STEM images and the Ti-L2,3spectra at
the interface look very similar. In particular, the
very similar peak splittings at ~462 eVin the Ti L2
edgessuggestthattheelectrontransferfromtheRO
layertotheneighboringTiO2planesisthesamefor
both LaO- and SmO-based heterostructures. Our
transportmeasurementsindicatethattheseelectrons
produce a conducting 2DEG in LaO heterostruc-
tures but are not mobile in SmO heterostructures.
TiO6octahedra rotations in bulk RTiO3deter-
mine the width of the Ti-3d band of t2gsym-
metry,andhencetheelectronicproperties,through
a change in the Mott-Hubbard gap (23). SrTiO3,
however,hasnoTiO6octahedralrotationsatroom
temperature. We investigated octahedral rotations
inourSrTiO3/RO/SrTiO3heterostructures, with
synchrotron x-ray experiments at the Advanced
Photon Source. We observed strong superlattice
reflections (figs. S7 and S8) resulting from unit-
cell doubling TiO6octahedra rotations, in good
agreement with the density functional calcula-
tions discussed below. The octahedral rotations
are well ordered in the interfacial plane, with typi-
cal rocking widths giving an in-plane domain
size > 60 nm. The breadths of the half-order
peaks in the out-of-plane direction are consistent
with octahedral rotations at the RTiO3layer
rapidly decaying into the SrTiO3matrix. These
decaying octahedra rotations lead to a spatial
gradient in the electronic structure, influencing
the conduction.
In addition, epitaxial strain in the interfacial
RTiO3layer also affects the interface conductiv-
ity. LaTiO3, PrTiO3, and NdTiO3layers at the
interface are strained under biaxial compression,
but SmTiO3and YTiO3layers are under biaxial
tension (table S1) (21). Compressive strain has
been shown to induce conducting behavior in
LaTiO3thin films (24), attributed to an increased
Ti t2gbandwidth and a weakened crystal field.
This has been predicted theoretically to reduce
the effect of electron correlations and to support
metallic behavior (25). The tensile strain in the
SmTiO3and YTiO3layers embedded in SrTiO3
appears to enforce the effect of strong correla-
tions and favor insulating behavior.
To understand the combined effects of charge
transfer, spatially varying octahedral rotations,
biaxial strain, and rare-earth electronic structure,
we have performed density functional theory
(DFT) calculations, including a Hubbard U term
accounting for the on-site Coulomb interaction
(20). The values of U that provide a realistic
description of the electronic and atomic structure
ofbulkYTiO3andLaTiO3compounds(26)were
used. The atomic positions were fully relaxed,
under the constraint that the in-plane lattice
constantbeequaltothecalculatedlatticeconstant
of bulk SrTiO3. The density of electronic states,
and the corresponding atomic coordinates,
calculated for periodic superlattices, are shown
inFig.4A(3.5-ucSrTiO3/1-MLLaO)andinFig.
4B (3.5-uc SrTiO3/1-ML YO). For the LaO-
basedheterostructure,theFermienergyliesinthe
region of nonzero density of states, consistent
with the previous calculations (27, 28), whereas
for the YO heterostructure the Fermi energy
lies between the split-off lower Hubbard band
and the higher energy density of states. This in-
dicates that the LaO-based interface is metallic,
whereas the YO-based interface is insulating,
supporting our experimental observations. Our
calculations predict that the ground state of
the SrTiO3/LaO heterostructure is not charge-
ordered, whereas the SrTiO3/YO heterostructure
is unstable with respect to charge dispropor-
tionation and has a charge-ordered ground state
similar to that found in (29). Octahedra rotations
are clearly visible in the relaxed structuresshown
in Fig. 4, C and D, consistent with our synchro-
tron measurements.
The electron donated by the RO embedded in
the SrTiO3matrix is localized to the nearby TiO2
layers. Filling of the Ti-3d band in these layers
close to n = 0.5, and enhanced electron-
correlation effects due to 2D confinement, will
strongly influence the interfacial conductivity.
It is well known that the effect of Ti-3d band
filling on electronic, magnetic, and transport
properties of bulk RTiO3Mott-Hubbard insula-
tors depends critically on the rare-earth ion (23).
It appears that for the relatively weakly corre-
lated LaO-based heterostructure, several percent
of hole doping is sufficient to cause a metal-
insulator transition. In contrast, for the YO-based
heterostructures with larger U, lower bandwidth
W, and larger strain and structural distortions,
the insulating phase persists. The number of elec-
trons transferred in each case is the same, but
strongercorrelationeffectsintheYOheterostruc-
ture seem to be responsible for the insulating
behavior. Our experimental and theoretical in-
vestigationssuggest that these correlations arise
from an interplay of strain, spatially varying ro-
tational distortions, and rare-earth ion effects on
the band structure. Indications of electron cor-
relations have also been recently reported in
LaIO3/SrTiO3heterostructures (30).
Strong correlations in 2DEGs at oxide inter-
faces have been shown to result from electronic
properties of different RO inserted layers, as well
as the structural and electronic modification of
nearbylayers.Quantitativelyexploringtheunder-
lying physics of the experimental data presented
here is complex and challenging, because strong
correlationscombinedwithatomic-scalestructural
andchemicalvariationsseverelylimittheeffective-
ness of theoretical calculations. The details can-
notbefullycapturedwithintheDFT+Ucalculations
used here, and more advanced approaches—
based on dynamical mean-field theory (31), for
example—are likely necessary to capture the spa-
tial variations. The work presented here is impor-
tant in elucidating correlation effects in systems
with atomic-scale perturbations (32) and external
perturbation-induced changes inoxide 2DEGsys-
tems (8, 15–17). The ability to design and grow
heterostructureswithatomic-scalevariations,and
the demonstrated strong dependence of correlated
2DEGs on these variations, open new directions
for oxide 2DEG heterostructures.
References and Notes
1. H. Yamada et al., Science 305, 646 (2004).
2. A. Ohtomo, D. A. Muller, J. L. Grazul, H. Y. Hwang,
Nature 419, 378 (2002).
3. M. P. Warusawithana, E. V. Colla, J. N. Eckstein,
M. B. Weissman, Phys. Rev. Lett. 90, 036802
(2003).
4. E. Bousquet et al., Nature 452, 732 (2008).
5. M. P. Warusawithana et al., Science 324, 367
(2009).
Fig. 4. Energy-dependent
densityofstatesandstruc-
tural relaxation of 3.5-uc
SrTiO3/1-ML LaO (A and
C)periodicsuperlatticeand
3.5-uc SrTiO3/1-ML YO pe-
riodic superlattice (B and
D)obtainedfromDFTcal-
culations. Positive density
ofstatesisforspinupand
negative is for spin down.
The dashed line indicates
the position of the Fermi
level. The results indicate
conducting behavior for
the 3.5-uc SrTiO3/1-ML
LaOperiodicsuperlattice
and insulating behavior
for the 3.5-uc SrTiO3/1-ML
YO periodic superlattice.
18 FEBRUARY 2011 VOL 331
SCIENCE
www.sciencemag.org
888
REPORTS

Page 4

6. V. Garcia et al., Nature 460, 81 (2009).
7. H. Kroemer, Surf. Sci. 132, 543 (1983).
8. J. Mannhart, D. G. Schlom, Science 327, 1607
(2010).
9. A. Ohtomo, H. Y. Hwang, Nature 427, 423 (2004).
10. N. Nakagawa, H. Y. Hwang, D. A. Muller, Nat. Mater. 5,
204 (2006).
11. Y. Hotta, T. Susaki, H. Y. Hwang, Phys. Rev. Lett. 99,
236805 (2007).
12. A. Brinkman et al., Nat. Mater. 6, 493 (2007).
13. N. Reyren et al., Science 317, 1196 (2007).
14. Y. Kozuka et al., Nature 462, 487 (2009).
15. S. Thiel, G. Hammerl, A. Schmehl, C. W. Schneider,
J. Mannhart, Science 313, 1942 (2006).
16. C. Cen et al., Nat. Mater. 7, 298 (2008).
17. C. Cen, S. Thiel, J. Mannhart, J. Levy, Science 323, 1026
(2009).
18. S. Okamoto, A. J. Millis, Nature 428, 630 (2004).
19. Y. Okimoto, T. Katsufuji, Y. Okada, T. Arima, Y. Tokura,
Phys. Rev. B 51, 9581 (1995).
20. Materials and methods are available as supporting
material on Science Online.
21. H. D. Zhou, J. B. Goodenough, J. Phys. Condens. Matter
17, 7395 (2005).
22. K. Janicka, J. P. Velev, E. Y. Tsymbal, Phys. Rev. Lett. 102,
106803 (2009).
23. M. Imada, A. Fujimori, Y. Tokura, Rev. Mod. Phys. 70,
1039 (1998).
24. F. J. Wong et al., Phys. Rev. B 81, 161101 (2010).
25. H. Ishida, A. Liebsch, Phys. Rev. B 77, 115350
(2008).
26. S. Okatov, A. Poteryaev, A. Lichtenstein, Europhys. Lett.
70, 499 (2005).
27. Z. S. Popovic, S. Satpathy, Phys. Rev. Lett. 94, 176805
(2005).
28. S. Okamoto, A. J. Millis, N. A. Spaldin, Phys. Rev. Lett.
97, 056802 (2006).
29. R. Pentcheva, W. E. Pickett, Phys. Rev. Lett. 99, 016802
(2007).
30. M. Breitschaft et al., Phys. Rev. B 81, 153414 (2010).
31. G. Kotliar et al., Rev. Mod. Phys. 78, 865 (2006).
32. Q. Si, M. J. Rozenberg, G. Kotliar, A. E. Ruckenstein,
Phys. Rev. Lett. 72, 2761 (1994).
33. We thank D. G. Schlom and D. A. Muller for fruitful
discussions. This work was supported by the National
Science Foundation under grant DMR-0906443 and a
David and Lucile Packard Fellowship (C.B.E.). The
research at University of Nebraska was supported by
the Materials Research Science and Engineering Center
(NSF grant DMR-0820521), the Nanoelectronics
Research Initiative of the Semiconductor Research
Corporation, the National Science Foundation (grant
EPS-1010674), and the Nebraska Research Initiative.
Work at the University of Michigan was supported
by the U.S. Department of Energy (DOE) under grant
DE-FG02-07ER46416. We thank the National Center for
Electron Microscopy at Lawrence Berkeley National
Laboratory for their support under DOE grant
DE-AC02-05CH11231 for user facilities. Work at Argonne
and use of the Advanced Photon Source were
supported by the DOE Office of Science, Office of
Basic Energy Sciences, under contract DE-AC02-
06CH11357. Work at Brookhaven National Laboratory
was sponsored by DOE/BES/MSE and the Center for
Functional Nanomaterials under contract DE-AC02-
98CH10886. J. Karapetrova’s assistance at beamline
33-BM of the Advanced Photon Source is gratefully
acknowledged.
Supporting Online Material
www.sciencemag.org/cgi/content/full/331/6019/886/DC1
Materials and Methods
Figs. S1 to S8
Table S1
References
7 October 2010; accepted 19 January 2011
10.1126/science.1198781
Time-Reversed Lasing and
Interferometric Control of Absorption
Wenjie Wan, Yidong Chong, Li Ge, Heeso Noh, A. Douglas Stone, Hui Cao*
In the time-reversed counterpart to laser emission, incident coherent optical fields are perfectly
absorbed within a resonator that contains a loss medium instead of a gain medium. The incident
fields and frequency must coincide with those of the corresponding laser with gain. We demonstrated
this effect for two counterpropagating incident fields in a silicon cavity, showing that
absorption can be enhanced by two orders of magnitude, the maximum predicted by theory
for our experimental setup. In addition, we showed that absorption can be reduced substantially
by varying the relative phase of the incident fields. The device, termed a “coherent perfect
absorber,” functions as an absorptive interferometer, with potential practical applications in
integrated optics.
T
chanics. It implies that if a particular physical
process is allowed, then there also exists a “time-
reversed process” that is related to the original
process by reversing momenta and the direction
ofcertainfields(typicallyexternalmagneticfields
and internal spins). These symmetry operations
are equivalent to changing the sign of the time
variableinthedynamicalequations,andforsteady-
state situations they correspond to interchanging
incoming and outgoing fields.
The power of time-reversal symmetry is that
it enables exact predictions of the relationship
betweentwoprocessesofarbitrarycomplexity.A
familiarexampleisspinechoinnuclearmagnetic
resonance (NMR) (1): A set of precessing spins
in a magnetic field fall out of phase because of
ime-reversal symmetry is a fundamental
symmetry of classical electromagnetic
theory and of nonrelativistic quantum me-
slightlydifferentlocalfieldenvironments,quench-
ing the NMR signal. The signal can be restored
by imposing an inversion pulse at time T, which
has the effect of running the phase of each spin
backwardintime,sothatafter2Ttheyarebackin
phase,nomatterhowcomplicatedtheirlocalfield
environment. Time-reversal symmetry is the ori-
gin of the well-known weak localization effect
(2) in the resistance of metals, the coherent back-
scattering peak in the reflection from multiple
scattering media (3–5), and the elastic enhance-
ment factor familiar in nuclear scattering (6). Ef-
fects due to direct generation of time-reversed
waves via special “mirrors” have been extensive-
ly studied for sound waves (7–9) and microwave
radiation (10).
Recently, several of the authors (11) explored
theoreticallyanexacttime-reversalsymmetryprop-
erty of optical systems: the time-reversed analog
of laser emission. In the lasing process, a cavity
with gain produces outgoing optical fields with a
definite frequency and phase relationship, with-
outbeingilluminatedbycoherentincomingfields
at that frequency. The laser is coupled to an en-
ergy source (the pump) that inverts the electron
populationofthegainmedium,causingtheonset
Department of Applied Physics, Post Office Box 208284, Yale
University, New Haven, CT 06520, USA.
*To whom correspondence should be addressed. E-mail:
hui.cao@yale.edu
Fig. 1. A laser beam from a
tunable (800 to 1000 nm) con-
tinuous-wave Ti:sapphire source
enters a beam splitter (desig-
nated 1). The two split beams
are directed normally onto oppo-
site sides of a silicon wafer of
thickness~110mm,usingaMach-
Zehnder geometry. A phase de-
lay in one of the beam paths
controlstherelativephaseofthe
two beams. An additional atten-
uatorensuresthattheinputbeams
have equal intensities, compen-
satingforimbalancesinthebeam
splitters and other imperfections. The output beams are rerouted, via beam splitters (designated 2, 3, and
4), into a spectrometer. The inset is a schematic of the CPA mechanism. The incident beams from left and
rightmultiplyscatterwithinthewaferwithjusttherightamplitudeandphasesothatthetotaltransmitted
and reflected beams destructively interfere on both sides, leading to perfect absorption.
www.sciencemag.org
SCIENCE
VOL 33118 FEBRUARY 2011
889
REPORTS

Page 5

www.sciencemag.org/cgi/content/full/331/6019/886/DC1




Supporting Online Material for

Metallic and Insulating Oxide Interfaces
Controlled by Electronic Correlations
H. W. Jang, D. A. Felker, C. W. Bark, Y. Wang, M. K. Niranjan, C. T. Nelson, Y. Zhang,
D. Su, C. M. Folkman, S. H. Baek, S. Lee, K. Janicka, Y. Zhu, X. Q. Pan, D. D. Fong,
E. Y. Tsymbal, M. S. Rzchowski, C. B. Eom*
*To whom correspondence should be addressed. E-mail: eom@engr.wisc.edu
Published 18 February 2011, Science 331, 886 (2011)
DOI: 10.1126/science.1198781


This PDF file includes:

Materials and Methods
Figs. S1 to S8
Table S1
References

Page 6

S-2

Materials and methods

Growth
All heterostructures were grown on TiO2-terminated SrTiO3 substrates by pulsed laser
deposition (PLD) in an oxygen atmosphere with in-situ reflection high-energy electron
diffraction (RHEED). Two types of heterostructures were grown. In the first, we deposited a RO
monolayer from a R2O3 target, followed by deposition of a SrTiO3 overlayer. In the second we
directly deposited a RTiO3 unit cell from a RTiO3+x target, followed by deposition of a SrTiO3
overlayer. Both types of heterostructures were grown with five different rare-earth ions (R = La,
Pr, Nd, Sm, Y), and with various SrTiO3 overlayer thicknesses varying from 0 to 40 unit cells.
Heterostructures of both types with R = La were grown at various oxygen pressures from 10-6
mbar to 0.1 mbar at a fixed growth temperature of 500 oC, and growth temperatures from 500 oC
to 800 at a fixed growth pressure of 10-3 mbar to determine optimal growth parameters of T =
500–550 oC and
2
O
P = 10-3 mbar. The laser energy density of 2 J/cm2 and the repetition rate of
2–3Hz were used. The distance from the target surface to the sample was 55 mm and the growth
rate of the film was 25–35 pulses per unit cell.

Electrical measurements
Electrical characteristics of most samples were measured with the van der Pauw technique.
Ohmic contacts onto four corners of a 5 mm × 5 mm sample were formed using Al wedge
bonding directly connected to the sample surface. Comparison samples with wire bonding to ion-
beam etched, metalized contacts in a van der Pauw geometry did not show substantial difference.
Samples patterned in a Hall configuration also showed similar electrical characteristics. The data
presented in this report were extracted from van der Pauw measurements with leads wire bonded
directly to the sample surface. Nominal sheet carrier concentration
s n and carrier mobility µ
were determined as
Hs
eRHn/
=
and
€
µ = RH/HRo, where
€
RH is the Hall resistance,
€
Ro the
sheet resistance, H the magnetic field applied perpendicular to the substrate, and e the magnitude
of the electron charge. This assumes a single band of carriers, and a spherical Fermi surface.

Page 7

S-3
Figure S2A shows the dependence of nominal sheet carrier density (ns) and mobility (µ) on
growth oxygen pressure. (PO2) for 10 uc SrTiO3 / 1 ML LaO / SrTiO3 heterostructures from a
basic interpretation of the Hall measurements. The dependence of ns on PO2 is significant,
whereas µ is insensitive to PO2. The decrease of ns with increasing PO2 is consistent with the
previous reports for LaAlO3/SrTiO3 system (S1,S2,S3,S4,S5). For PO2 ≥ 10-3 mbar, carrier
concentrations as low as 1013 cm-2, which is believed to be an intrinsic feature of the conducting
LaO/TiO2 interface (S1,S5,S6,S7,S8), can be obtained without additional post annealing. In
contrast to PO2, we found that there is no notable dependence of ns and µ on growth temperature
(Fig. S1). A 10 uc SrTiO3 / 1 ML LaO / SrTiO3 heterostructure grown at PO2 = 10-3 mbar shows
almost constant ns from 3 K to 300 K in Fig. S2B. The measured µ of 1250 cm2V/s at 3 K is
similar to that previously reported for oxidized LaAlO3/SrTiO3 heterostructures (S5).

First principles calculations
First-principles electronic structure calculations have been performed within density
functional theory (S9) applied to 3.5 uc SrTiO3/1ML MO (M = La, Y) superlattices. The
calculations employed the projector augmented wave (PAW) method (S10) as implemented in
the Vienna Ab Initio Simulation Package (VASP) code (S11) The exchange-correlation effects
were treated within generalized gradient approximation (GGA). The calculations were carried
out using a plane-wave basis set limited by a cutoff energy of 520 eV and the 6×6×4 mesh of k
points in the irreducible Brillouin zone with energy converged to 10-5 eV/cell. Atomic
relaxations were performed until the Hellmann-Feynman forces on atoms have become less than
10 meV/Å. The lateral unit cells of the structures were constrained to have c(2×2) symmetry and
the in-plane lattice constant to be the calculated bulk lattice constant of SrTiO3.
Electron-electron correlations of the partially filled Ti 3d-states were treated within the
GGA+U approach (S12) that involves a Hubbard U term responsible for the on-site Coulomb
interaction. The values of U were chosen to provide a reasonable agreement between the
calculated and measured optical band gaps for bulk LaTiO3 and YTiO3 compounds. Thus, for the
LaO-based structure the value of U was chosen to be 3 eV, while for the YO-based structure the

Page 8

S-4
value of U was fixed at 4 eV. The calculated electronic structures of bulk LaTiO3 and YTiO3 are
in agreement with previous calculations (S13) and describe correctly the atomic structures and
the ground states of these compounds. The smaller U in the case of LaTiO3 as compared to
YTiO3 is justified (S14) since the bandwidth of the Ti 3d states in LaTiO3 is smaller than that in
YTiO3 due to larger GdFeO3-type distortions for the latter. This is found to be also the case for
the 3.5 uc SrTiO3/1ML MO superlattices where the octahedral distortions are much more
pronounced for M = Y than for M = La (see Figs. 4C, 4D). In particular, the tilting angles of the
Ti-O-Ti bonds along the [001] axis for the oxygen octahedra ending at the M atom are 147° for
M = Y as compared to 162° for M = La. Since the La 4f bands lie at higher energy than that
predicted by GGA, the value of U for the La 4f states was taken to be 8.0 eV to avoid their
spurious mixing with the conduction bands. The value of the exchange parameter was fixed at
J = 1 eV since its value is almost independent of the choice of materials.
Including octahedral distortions in the structural model is essential (S15) and leads to a
significant reduction in energy for the considered c(2×2) structure as compared to the p(1×1)
structure (S16). For the values of U chosen, we have that this energy difference is about 0.42 eV
per equivalent (1×1) lateral cell for the LaO-monolayer system and 1.38 eV for the YO-
monolayer system. Our calculations also showed that while the ground state of the SrTiO3/LaO
heterostructure is non-charge ordered, the SrTiO3/YO heterostructure is unstable with respect to
charge disproportionation and has a charged-ordered ground state (S17). Within the c(2×2)
structure constraint the latter is a ferromagnetic insulator characterized by two non-equivalent Ti
atoms in the plane of the structure with magnetic moments of 0.90 µB and 0.05 µB at the interface
layer and nearly zero magnetic moments away from the interface. The charge and orbital
ordering are similar to those found in ref. (S17) and characterized by charge disproportionation
Ti3.9+/ Ti3.05+ with preferential Ti3.05+ dxy-orbital occupation at the interface. The gain in energy
for the charge ordered configuration as compared to the non-charge ordered configuration for the
SrTiO3/YO system is about 0.14 eV per equivalent (1×1) lateral cell. For SrTiO3/MO
heterostructures with smaller SrTiO3 thickness of 1.5 uc we also checked a possibility of the
p(2×2) charge-ordered anti-ferromagnetic state. We found that the total energy of this state is
higher by about 0.11eV for the SrTiO3/LaO structure, while for the SrTiO3/YO structure it is
lower by 12 meV, however, the latter value is close to the accuracy of our computations.

Page 9

S-5

SrTiO3 overlayer critical thickness
Additional experimental insight comes from our measured dependence on the SrTiO3
overlayer thickness. We have found that although a LaO layer embedded in SrTiO3 is conducting,
a 1ML LaO film (d = 0 uc) on SrTiO3 is insulating. The film is still insulating with a 2 uc SrTiO3
overlayer, but becomes conducting with a 3 uc SrTiO3 overlayer, corresponding to the critical
thickness (dc) of 3 uc. We obtained similar results for different rare earth ions (see Fig. S3A and
S3B). We measured the ns of SrTiO3/1ML LaO films on SrTiO3 with increasing thickness of
SrTiO3 overlayer (d). For d ≥ 5 uc, the film shows a saturation in ns, which means that the
conductance is confined to a narrow thickness range of the sample, that is, a 2DEG. This could
be explained by a change in the Ti-3d band filling. In the absence of the SrTiO3 overlayer an
extra electron provided by the LaO is localized close to the surface making the band filling of the
interfacial TiO2 monolayer close to n = 1 as in bulk insulating LaTiO3.
Depositing SrTiO3 on top of LaO leads to a redistribution of electron density between the
top and bottom TiO2 layers, reducing the Ti 3d band occupation and resulting in the transition to
a metallic state, as discussed previously in a different context. Our calculations predict, however,
that one unit cell of the SrTiO3 overlayer is sufficient to cause a metal-insulator transition in the
system, somewhat inconsistent with our 3 uc experimental SrTiO3 critical thickness. This may be
due to limitations of the DFT calculations, or to the presence of adsorbates on the surface
neglected in our structural model. Alternatively, surface adsorbates providing low-energy states
could pull electrons from the RO monolayer, producing insulating behavior for small (or zero)
SrTiO3 overlayer thicknesses. The electric field from the resulting charge separation would raise
the surface state energies with increasing SrTiO3 thickness, leading to repopulation of the RO
interface above a critical SrTiO3 thickness.

STEM and EELS analysis
Transmission electron microscopy studies were performed using an aberration corrected
scanning transmission electron microscope (STEM), Hitachi HD2700C, equipped with a cold-
field-emission electron source and a high-resolution electron energy-loss spectrometer. Cross-
sectional cuts of the samples grown under the optimum condition described above were prepared

Page 10

S-6
by mechanical polishing followed by low-energy and low-angle ion milling. For high angle
annular dark field (HAADF) imaging, a probe size of 0.8–1 Å, a convergence angle of 28 mrad,
and a HAADF collection angle of 114–608 mrad were used. For electron energy-loss
spectroscopy (EELS) in STEM, a probe size of 1.3 Å, a convergence angle of 28 mrad, a
HAADF collection angle of 45–242 mrad, and an EELS collection angle of 20 mrad were used.
For a beam current of 50–100 pA the acquisition time for EELS is about 1-2 sec and the energy
resolution is about 0.4 eV.
The challenge for atomically resolved EELS study of the heterostructures is to minimize
radiation damage on the sample. Our study shows that the electron illumination degrades the
valence state of Ti from Ti4+ to Ti3+. Therefore, in this study, to obtain good signal-noise ratio
spectra and avoid the artifact from the beam damage, we used the selected area EELS and 2D
scanning methods with proper probe currents (around 50 pA) and scanning step (≥ the probe
size). The quality of the data were monitored from the HAADF image (Fig. S4) as well as from
the EELS spectra, which were then compared with those at the SrTiO3 substrate far away from
the interface to ensure the spectra acquired are damage-free. The line scanning spacing between
each spectrum is 1.2–1.6Å. The 2D EELS scanning spectra presented in the manuscript were
selected every other spectrum from the raw data. The selected area EELS spectra of Ti L2,3 edge
were normalized by the peak height of the Ti L2 edges.
Thick films of the reference samples of LaTiO3 (Ti3+) and SrTiO3 (Ti4+) were prepared
and measured. The EELS data was analyzed using the method described in Ref. 2 of the main
text. The experimental spectrum at the LaO layer was fit with a linear combination of the Ti L2,3
edges of LaTiO3 and SrTiO3. The analysis based on these data shown in Fig. 4B of our
manuscript indicates that the fractional contribution at the LaO layer is 24% for Ti3+ while 76%
for Ti4+. The decay of the Ti3+ and La signal from the LaO layer into the SrTiO3 film was fit with
the Lorentzian distribution, yielding an ~1 nm spatial distribution (full width at half maximum)
of Ti3+ and La across the LaO layer (see Fig. S5, S6)

Synchrotron diffraction measurements
Room temperature synchrotron x-ray scattering measurements of the 10 uc SrTiO3/1 ML
LaO/SrTiO3 and 10 uc SrTiO3/1 ML SmO/SrTiO3 heterostructures were performed at Sector 33-

Page 11

S-7
BM of the Advanced Photon Source. In-plane (K) scans along the [010] direction are shown in
Fig. S6, where reciprocal lattice units are referenced to the SrTiO3 lattice parameter at room
temperature. As seen, both samples exhibit sharp half-order peaks, indicating the presence of a
structure with 2×2×2 unit cell periodicity with respect to the SrTiO3 substrate, as expected for
perovskites with tilted oxygen octahedra (S18). While ½{odd, odd, odd} peaks were observed
for both H ≠ K and H or K ≠ L (such as ½{3, 1, 1} and ½{3, 3, 1}, respectively), mixed odd and
even integer peaks such as the ½{3, 0, 1} were absent, as were peaks of the form ½{H, H, H}.
Following the structure factor calculations of Ref. S10, this allows us to conclude that the Glazer
tilt system is a-a-c-, where we have assumed in-plane, biaxial symmetry. This is in good
agreement with the results of computational theory. Half-order reflections were not observed
from the SrTiO3 substrate, which would produce strong, sharp peaks in positions consistent with
bulk SrTiO3.
In-plane rocking curves taken on the ½{3, 1, 1} peaks for the 10 uc SrTiO3 / 1 ML LaO/
SrTiO3 and 10 uc SrTiO3 / 1 ML SmO / SrTiO3 heterostructures are shown in Figure S7(a). From
the FWHMs, we find the in-plane domain size is ≥ 88 nm for the LaO sample and ≥ 60 nm for
the SmO sample. Scans along L, the out-of-plane direction, are shown in Figure S7(b). For both
samples, the half-order reflections are similarly broad along L, with the FWHM corresponding to
a domain width of ~2 nm.

Page 12

S-8
A
B
C
10-7
10-6
10-5
10-4
10-3
10-2
10-1
100
1012
1013
1014
1015
1016
1017
10 uc SrTiO3/1ML LaO/SrTiO3
10 uc SrTiO3/1 uc LaTiO3/SrTiO3
Carrier concentration (cm-2)
Growth oxygen pressure (mbar)
400500
Growth temperature (oC)
600700 800 400 500
Growth temperature (oC)
600 700800
Temperature (K)
100
101
Temperature (K)
102
10 uc SrTiO3/1ML LaO/SrTiO3
10 uc SrTiO3/1 uc LaTiO3/SrTiO3
10-7
10-6
10-5
10-4
10-3
10-2
10-1
100
Growth oxygen pressure (mbar)
10-1
100
101
102
103
104
Mobility (cm2/Vs)
1012
1013
1014
1015
1016
1017
Carrier concentration (cm-2)
10-1
100
101
102
103
104
Mobility (cm2/Vs)
1012
1013
1014
1015
1016
1017
Carrier concentration (cm-2)
10-1
100
101
102
103
104
Mobility (cm2/Vs)
10 uc SrTiO3/1ML LaO/SrTiO3
10 uc SrTiO3/1 uc LaTiO3/SrTiO3
10 uc SrTiO3/1ML LaO/SrTiO3
10 uc SrTiO3/1 uc LaTiO3/SrTiO3
10 uc SrTiO3/1ML LaO/SrTiO3
10 uc SrTiO3/1 uc LaTiO3/SrTiO3
10 uc SrTiO3/1ML LaO/SrTiO3
10 uc SrTiO3/1 uc LaTiO3/SrTiO3
100
101
102

Figure S1. Electrical properties of 10 uc SrTiO3/1ML LaO/SrTiO3 and 10 uc SrTiO3/1 uc
LaTiO3/SrTiO3 heterostructures. (A) Sheet carrier concentration and mobility as a function of
growth oxygen partial pressure. The growth temperature was 500 oC. (B) Sheet carrier
concentration and mobility as a function of growth temperature. The growth oxygen pressure
was 10-3 mbar. (C) Sheet carrier concentration and mobility as a function of temperature. The
growth temperature and oxygen pressure were 500 oC and 10-3 mbar, respectively.

Page 13

S-9
A?
2?
10-6? 10-5? 10-4? 10-3? 10-2? 10-1?
PO (mbar)?
1013?
1014?
1015?
1016?
!?
ns (cm-2)?
10-1?
100?
101?
102?
T = 300 K?
ns?
µ?
µ (cm2/Vs)?


B?
!?
T (K)?
ns (cm-2)?
1012?
1013?
1014?
1015?
1011?
100?
101?
102?
101?
102?
103?
104?
100?
ns?
µ?
µ (cm2/Vs)?
C?
5?
10?
d (uc)?
15?
20?
1013?
1014?
ns (cm-2)?
T = 300 K?
1011?
1012?
0?
1010?
102?
103?
100?
101?
10-1?
ns?
µ?
µ (cm2/Vs)?

Figure S2. Electrical properties of SrTiO3/1ML LaO/SrTiO3 heterostructures. (A) Sheet carrier
concentration and mobility as a function of growth oxygen pressure for 10 uc SrTiO3/1ML
LaO/SrTiO3 heterostructures. The horizontal dotted line indicates a plateau carrier concentration
in the intermediate pressure regime, indicating that the carrier concentration does not depend on
oxygen pressure in this regime. (B) Sheet carrier concentration and mobility as a function of
temperature for a 10 uc SrTiO3/1ML LaO/SrTiO3 heterostructure grown at PO2= 10-3 mbar. (C)
Sheet carrier concentration and mobility as a function of the thickness of SrTiO3 overlayer for
SrTiO3/1ML LaO/SrTiO3 heterostructures grown at PO2=10-3 mbar. The measurement limit in
sheet carrier concentration indicated by the horizontal dashed line is 1010 cm-2.

Page 14

S-10

B
C
D
Carrier concentration (cm-2)
100200300
10 uc SrTiO3/1 uc LaTiO3/SrTiO3
10 uc SrTiO3/1 uc PrTiO3/SrTiO3
20 uc SrTiO3/1 uc NdTiO3/SrTiO3
40 uc SrTiO3/1 uc NdTiO3/SrTiO3
Temperature (K)
Carrier concentration (cm-2)
1011
1012
1013
1014
1015
Carrier concentration (cm-2)
A
Carrier concentration (cm-2)
1010
1011
1012
1013
1014
1015
108
109
40 uc SrTiO3/1ML RO/SrTiO3
40 uc SrTiO3/1 uc RTiO3/SrTiO3
LaPrNdSmY
R ion
1010
Thickness of SrTiO3 overlayer (uc)
5101520253035400
1011
1012
1013
1014
1010
LaO
PrO
NdO
SmO
YO
LaTiO3
PrTiO3
NdTiO3
SmTiO3
YTiO3
Thickness of SrTiO3 overlayer (uc)
5101520253035400
1011
1012
1013
1014
1010


Figure S3. Dependence of sheet carrier concentration on the R (= La, Pr, Nd, Sm, Y) ion in
SrTiO3/1ML RO/SrTiO3 and SrTiO3/1 uc RTiO3/SrTiO3 heterostructures. (A) Sheet carrier
concentration as a function of the thickness of the SrTiO3 overlayer in SrTiO3/1ML RO/SrTiO3
heterostructures. (B) Sheet carrier concentration as a function of the thickness of the SrTiO3
overlayer in SrTiO3/1 uc RTiO3/SrTiO3 heterostructures. (C) Sheet carrier concentration as a
function of RO or RTiO3 doping layers. (D) Temperature dependence of sheet carrier
concentration in conducting SrTiO3/1 uc RTiO3/SrTiO3 heterostructures. NdTiO3-based
heterostructures show semiconducting behavior, whereas LaTiO3- and PrTiO3-based
heterostructures show metallic behavior.

Page 15

S-11

























Figure S4. HAADF images of (A) 10 uc SrTiO3/1ML LaO/SrTiO3, (B) 10 uc SrTiO3/1ML
SmO/SrTiO3, (C) 10 uc SrTiO3/1 uc LaTiO3/SrTiO3, and (D) 10 uc SrTiO3/1 uc SmTiO3/SrTiO3
heterostructures. The arrows indicate the interfacial LaO or SmO layers.




A
C
B
D