1

Piezoforce Iagig of Cofied Oxide Naowires

Akash Levy
Allderdice High School, Pittsburgh, PA 15217
akashlevy@gmail .com

Abstract
A technique called piezoresponse force microscopy (PFM) was used to non-destructively image
and dimension conductive nanostructures created at the interface between LaAlO3 and SrTiO3.
Comparisons of PFM imaging with a previously established destructive method for quantifying ~10
nanometer nanowires indicate very good agreement.
Nanoscale regions at the interface of LaAlO3/SrTiO3 can be reversibly switched between
electrically conducting and electrically insulating with an Atomic Force Microscope (AFM) tip. A software
program was developed to scan a voltage-biased AFM tip in a programmed pattern over the LaAlO3
surface, allowlng nanoscale llnes and oLher complex paLLerns Lo be wrlLLen" and erased" aL Lhe
LaAlO3/SrTiO3 interface. A three-dimensional visualization tool was developed to monitor the progress
of the nanoscale writing. Simultaneous measurements of the conductivity of these nanostructures were
performed using a lock-in amplifier.
The PFM response of the LaAlO3/SrTiO3 interface was measured by applying an alternating
voltage to the conductive nanostructures, and measuring the resulting deflection of an electrically-
isolated AFM tip that scanned the surface of the sample. The raster-scanned PFM images were analyzed
to determine the wire widths. The PFM-derived widths were compared to those obtained from erasure
experiments. PFM had a spatial resolution of approximately 25 nm, with the advantage of being non-
destructive.
PFM imaging may become a useful technique for researchers interested in developing new
families of nanoelectronics circuitry using the LaAlO3/SrTiO3 interface. The technique provides a
relatively simple, high-resolution method for the non-destructive imaging of conductive nanostructures.
1 mm
2

Contents
Abstract ......................................................................................................................................................... 1
Introduction .................................................................................................................................................. 3
Development of Software for AFM Lithography........................................................................................... 4
Visualization and Measurement of Nanostructures ..................................................................................... 5
Procedure ...................................................................................................................................................... 6
Results ........................................................................................................................................................... 8
Discussion.................................................................................................................................................... 10
Conclusions ................................................................................................................................................. 11
Acknowledgements ..................................................................................................................................... 11
References .................................................................................................................................................. 11


3

Introduction
In 2004, Ohtomo and Hwang
1
reported the existence of a two-dimensional electron gas (2DEG) at the
interface between two oxide semiconductors, LaAlO3 and SrTiO3. A 2DEG can be described as a layer of
electrons that are free to move in only two dimensions. Since 2DEGs were discovered in LaAlO3/SrTiO3,
their properties in these materials have been explored by many research groups. Many remarkable
phenomena have been discovered such as electrical conductivity
3
, magnetism
2
and superconductivity
4
.
For these phenomena to appear, the LaAlO3/SrTiO3 samples must be grown under very precise
conditions
5
(see structure in Figure 1(a)). It is crucial that the SrTiO3 is TiO2-terminated, otherwise many
of the maLerlals lnLeresLlng properties cannot be observed. The conductivity at the interface between
the LaAlO3 layer and the SrTiO3 layer is dependent on the thickness of the LaAlO3 layer
2
; at or below a
critical thickness of three unit cells of LaAlO3, the interface is insulating, while at or above four unit cells
thickness, the interface is conducting. Near the critical thickness, the conductivity of the sample can be
locally and reversibly controlled with a voltage-biased conductive atomic force microscope (c-AFM) tip.
Applying a positive voltage to a c-AFM tip in contact with the sample produces local conductivity at the
point of contact. Conversely, applying a negative voltage to a point on a sample restores the insulating
state at that point on the sample. By controlling the path and voltage of the c-AFM tip, it becomes
convenient to write nanowires onto the sample
6
.

Figure 1. Images of LaAlO3/SrTiO3. (a) A visualization of the molecular structure of LaAlO3/SrTiO3. (b) Nanowires being written on
a sample by a c-AFM tip.
The c-AFM lithography process is analogous to the sketching of a picture. The c-AFM tip acts as a pen",
the sample behaves as a sheeL of paper" and Lhe voltage applied acts as lnk". 1he plcLures" that are
sketched onto the sample are usually circuits and their components. It is possible to write nanodevices
such as field-effect transistors
6
(FETs), photodetectors
7
, terahertz sources and detectors
8
and single
electron transistors
9
(SETs) as depicted in Figure 2. All of these devices have potential applications in a
variety of disciplines. Field-effect transistors serve as the building block for nanoscale computational
systems. Terahertz microscopy has the potential to characterize single molecules placed between a
(LaO)
+
(AlO
2
)

(LaO)
+
(AlO
2
)

(LaO)
+
(TiO
2
)
0
(SrO)
0
(TiO
2
)
0
(SrO)
0
(TiO
2
)
0
(SrO)
0
(a) (b)
4

source and a detector. Although they have not yet been commercialized, room-temperature single
electron transistors could one day serve as low-power alternatives to the classical transistors that
operaLe ln Lhe ma[orlLy of Lodays elecLronlc devlces.

Figure 2. Examples of nanodevices produced by c-AFM lithography of LaAlO3/SrTiO3.
Development of Software for AFM Lithography
erformlng AlM llLhography requlres preclse conLrol over Lhe AlM Llps poslLlon and applled volLage.
Using LabVIEW in conjunction with Inkscape, this process becomes very manageable
13
. The software
used previously to perform lithography was very limited in functionality. It lacked important features
such as the ability to draw curved paths and fill in shapes. Thus, the software was redesigned from
scratch to enhance its capabilities (see Figure 3). The new program was modular, allowing it to operate
with several varieties of AFMs, and was also much more stable. Additionally, a simulation tool was
wrlLLen ln vyLhon LhaL operaLed as vlrLual hardware", offerlng Lhe ablllLy Lo LesL nanowrlLlng
procedures before conducting experiments. In addition, the tool provided a way to track the
nanowriting process during the experiments.
Photodetector Field effect transistor (FET) Single electron transistor (SET)
(a) (b) (c)
5


Figure 3. Front panel of the LabVIEW AFM lithography software. (RED) Mode, AFM Type, and Simulator controls. (ORANGE)
Controls to specify background image and circuit design files. (YELLOW) Properties of shapes to draw on sample surface.
(GREEN) A control that allows you to specify the order you want each of the shapes to be written. (BLUE) An estimated time
required for lithography based on the path lengths and speeds for the paths.

Visualization and Measurement of Nanostructures
One challenge of working with LaAlO3/SrTiO3 has been measuring the dimensions of nanowires
produced by c-AFM lithography. The written nanostructures cannot be imaged by an atomic force
microscope, because there is no measureable difference in the topography of a conducting versus an
insulating one. This essentially means that an AFM image of a sample with nanowires present appears
the same as an image of a sample without nanowires. One research group approached this problem
using electric force microscopy
10
(EFM). EFM imaging involves scanning the charge present at the
surface of a sample. Although this method was somewhat effective in visualizing nanostructures, it does
not directly measure the variations in carrier density that account for conductivity.
In the past, precise width measurement of nanowires has been performed by making cuts and
measuring the conductance drop as a function of time
11
. From the conductance drop, it is possible to
infer the width of that particular nanowire. However, this erasure method is not generally useful for
6

imaging nanostructures because it destroys the circuit. In this paper, a novel technique for imaging
samples of LaAlO3/SrTiO3 is reported. The discovery that a sample exhibits different piezoelectric
responses in conducting regions of the sample versus insulating regions allows for the use of piezoforce
microscopy (PFM) to do imaging. LaAlO3/SrTiO3 displays the inverse piezoelectric effect, meaning that
the material exerts an upward force on the tip in the presence of an electric field. The strength of this
upward force is measurably different for areas where the sample is conducting versus areas where it is
insulating, thus enabling a useful system for imaging nanostructures.

Figure 4. Methods of discovering nanostructure dimensions. (a) EFM images at various phases of nanowriting. (b) Erasure
experiment being conducted to measure wire width.
Procedure
The sample used was 35 m by 35 m in size. All experiments are carried out under ambient conditions
at room temperature. In the first part of the experiment, vlrLual elecLrodes" are creaLed by writing
rectangular-shaped areas at each of the gold electrodes with a tip voltage of +8 V. The use of virtual
electrodes maximizes electrical contact with the Au electrodes present on the sample. Next, a nanowire
is drawn between these electrodes with a tip voltage of +8 V at 600 nm/s. Then, the nanowire is cut
using a tip voltage of -10 V at 10 nm/s while the conductance drop is monitored. The conductance is
then fit to the curve () =
0

1
tanh() and the spatial derivative () is used to determine
the width of the wire. The wire width is calculated by finding the full-width at half-maximum (FWHM)
of () (see Figure 5 below).
(a) (b)
7


Figure 5. FWHM being calculated on a curve. The peak value of the curve, fmax is found then halved to find the half-maximum.
This value of f is used to obtain the two x values where the curve is at this half-maximum value. The distance between these
curves yields the full-width at half-maximum of the wire.
The next part is the PFM experiment. To begin, the resonant frequency of the AFM probe needs to be
found. This is important because at the resonant frequency, deflection of the AFM cantilever is
amplified, yielding a better signal-to-noise ratio. To find the resonant frequency, a sweep of a wide
range of frequencies is performed in search of a distinct peak. Once the resonant frequency is obtained,
PFM imaging can be conducted. With the AFM tip electrically isolated, the sample is scanned (as shown
in Figure 6) and the induced piezoresponse is measured through the vertical deflection of the AFM tip.
These piezoresponse data values can be used to create an intensity graph, which serves as a useful
qualitative image of the nanostructures present on the sample. Calculating the FWHM along each line of
the PFM image yields a set of values for the wire width. Averaging the FWHM values produces a value
for the wire thickness as measured by PFM.

Figure 6. PFM imaging involves raster scanning the region of sample where the nanowire is located while measuring the
piezoresponse as a function of position.
8

Results
Over several experiments, it was found that the resonant frequency of the AFM cantilever used is
between 275 kHz and 300 kHz as shown in Figure 7(a). The wire width measured in the erasure
experiment was 8 nm in width (Figure 8). The average FWHM measured in the PFM experiment yielded
a width of 33 3 nm (Figure 9(b) and (c)). After several experiments at varying writing speeds and tip
voltages, a linear regression was used (Figure 9(d)) to correlate the calculated width from the erasure
experiment (Wfit) and the width measured in the PFM experiment (WPFM). The line of best fit correlating
Wfit and WPFM yielded

= 1.u2

2S.6 nm. The R

2
value was 0.9175 and the p-value was
less than 0.01. This procedure can be used to characterize the properties of arbitrary nanostructures,
e.g., Figure 10.

Figure 7. Tuning the resonant frequency of the AFM cantilever and performing a PFM scan of a nanostructure. (a) Sweeping the
frequencies produces a maximum signal at approximately 283 kHz. This is the resonant frequency of the AFM cantilever. (b) A
PFM image of a nanostructure present at the interface.

Figure 8. Conductance curves as the nanowire is written and erased. (a) This graph plots the conductance G against time. The
first peak represents the conductance jump associated with the nanowire being written between the two electrodes. The drop
that occurs afterwards represents the conductance drop as the nanowire is being cut. (b) This graph plots the magnitude of the
(a) (b)
(a) (b)
9

change in conductance versus time. The FWHM measurement yields a precise value for the wire width that will later be
compared to the wire width obtained by PFM.

Figure 9. Results from the PFM experiment. (a) A PFM image of a single nanowire. The dashed line represents the y-value used
for the data points plotted in (b). (b) The data points represent the amplitude of the piezoresponse at given x-values along the
dashed line in (a). The best fit curve for the data is used to determine the FWHM. (c) Moving the dashed line in (a) along the y-
axis and recalculating the values plotted in (b) yields a set of FWHM values which are used to make a scatterplot. The average
width calculated from this scatterplot was approximately 33 nm with a 99% confidence interval from about 30 nm to 36 nm. (d)
This plot displays data from several experiments using different lithography parameters. A line of best fit is constructed to
correlate the widths found by PFM imaging with the widths found in erasure experiments.
30
25
20
15
10
5
0
W
f
i
t

(
n
m
)
55 50 45 40 35 30 25
W
PFM
(nm)
10V 200nm
10V 400nm
10V 600nm
10V 800nm
10V 1000nm
10V 1200nm
8V 200nm
8V 400nm
8V 600nm
8V 800nm
fit
(a) (b)
(c) (d)
10


Figure 10. Nanowriting and subsequent PFM imaging of a conductive nanostructure. (a) Screenshot showing AFM topography
with overlaid Inkscape pattern used to define the c-AFM path. (b) Three-dimensional visualization of nanowriting procedure. (c)
PFM image of the written nanostructure.

Discussion
The line of best fit yielded from the experiments that were conducted shows an approximately one-to-
one relationship between the measured line width and the precise line width as calculated from the
erasure experiments. The offset of 25.6 nm represents the size of the smallest feature that PFM imaging
can resolve. From the data obtained, it is also clear that writing speed and tip voltage play a role in
determining the width of a nanowire.
Although there was random error observed in repeated experiments, this error did not affect the
validity of the data significantly. The error may have been due to fluctuations in environmental factors
such as temperature and humidity or caused by failure to fully erase conducting regions between
experiments. However it was necessary to use the same sample for every experiment as each sample of
LaAlO3/SrTiO3 has slightly different characteristics.
14
12
10
8
6
4

m
14 12 10 8 6 4 2 0
m
-1.0
-0.5
0.0
0.5
V
(a) (b)
(c)
11

It was reasonable to hypothesize that the variation of the piezoelectric signal as a function of
conductivity at the interface would allow for an imaging technique. The point of interest was whether or
not this method of imaging was destructive and what maximum resolution is attainable. Piezoforce
imaging has been done on many other materials in the past but this is the first time it is being performed
to image nanostructures at the interface of LaAlO3 and SrTiO3. The resolution of the PFM images
produced is much higher than that of any other non-destructive imaging technique reported in the past.
The origin of the PFM signal is still not understood precisely. A previous study
12
showed that the writing
of nanowires causes structural reconstructions at the interface but it is not yet clear how these
reconstructions affect the piezoelectric properties of the sample. Further studies of these oxide
heterostructures could lead to an improved understanding of how structural reconfigurations change
the interfacial properties.
Conclusions
The results indicate that piezoforce microscopy delivers a high-resolution, non-destructive way to image
electrical nanostructures present at the interface of LaAlO3 and SrTiO3. PFM imaging can serve as a
technique for researchers studying nanostructures in LaAlO3/SrTiO3 to acquire the dimensions of
nanowires and the devices they form as well as provide qualitative data about the appearance and
function of a nanocircuit.
Acknowledgements
This work was carried out with Dr. Patrick Irvin and research group at the University of Pittsburgh. Figure
1(a) is adapted from Reference 2. Figure 2(a) is adapted from Reference 6, Figure 2(b) is adapted from
Reference 7, and Figure 2(c) is adapted from Reference 9. Figure 4(a) is adapted from Reference 10 and
Figure 4(b) is adapted from Reference 11. Figure 5 was illustrated by Arne Nordmann and published
under the Creative Commons Attribution-Share Alike 3.0 license. The resources used were supported by
grants from the U.S. National Science Foundation (NSF).
References
1. Ohtomo and H. Y. Hwang, A high-mobility electron gas at the LaAlO3/SrTiO3 heterointerface,
Nature 427(6973), 423 (2004).
2. S. Thiel, G. Hammerl, A. Schmehl, C. W. Schneider, and J. Mannhart, Tunable Quasi-Two-
Dimensional Electron Gases in Oxide Heterostructures, Science 313(5795), 1942 (2006).
3. A. Brinkman, M. Huijben, M. Van Zalk, J. Huijben, U. Zeitler, J. C. Maan, W. G. Van der Wiel, G.
Rijnders, D. H. A. Blank, and H. Hilgenkamp, Magnetic effects at the interface between non-
magnetic oxides, Nat. Mater. 6(7), 493 (2007).
4. N. Reyren, S. Thiel, A. D. Caviglia, L. F. Kourkoutis, G. Hammerl, C. Richter, C. W. Schneider, T.
Kopp, A. S. Ruetschi, D. Jaccard, M. Gabay, D. A. Muller, J. M. Triscone, and J. Mannhart,
Superconducting interfaces between insulating oxides, Science 317(5842), 1196 (2007).
5. J. W. Park, D. F. Bogorin, C. Cen, D. A. Felker, Y. Zhang, C. T. Nelson, C. W. Bark, C. M. Folkman, X.
Q. Pan, M. S. Rzchowski, J. Levy, and C. B. Eom, Creation of a two-dimensional electron gas at an
oxide interface on silicon, Nat. Commun. 1, 94 (2010).
6. C. Cen, S. Thiel, J. Mannhart, and J. Levy, Oxide Nanoelectronics on Demand, Science 323(5917),
1026 (2009).
12

7. P. Irvin, Y. Ma, D. F. Bogorin, C. Cen, C.W. Bark, C. M. Folkman, C.-B. Eom, and J. Levy, Rewritable
nanoscale oxide photodetector, Nat. Photon. 4(12), 849 (2010).
8. Y. Ma, M. Huang, S. Ryu, C. W. Bark, C. B. Eom, P. Irvin, and J. Levy, Broadband Terahertz
Generation and Detection at 10 nm Scale, Nano Lett. 13(6), 2884 (2013).
9. G. L. Cheng, P. F. Siles, F. Bi, C. Cen, D. F. Bogorin, C.W. Bark, C. M. Folkman, J.W. Park, C. B. Eom,
G. Medeiros-Ribeiro, and J. Levy, Sketched oxide single-electron transistor, Nat. Nanotechnol.
6(6), 343 (2011).
10. Y. W. Xie, C. Bell, T. Yajima, Y. Hikita, and H. Y. Hwang, Charge Writing at the LaAlO3/SrTiO3
Surface, Nano Lett. 10(7), 2588 (2010).
11. C. Cen, S. Thiel, G. Hammerl, C. W. Schneider, K. E. Andersen, C. S. Hellberg, J. Mannhart, and J.
Levy, Nanoscale control of an interfacial metalinsulator transition at room temperature, Nat.
Mater. 7(4), 298 (2008).
12. J. L. Maurice, C. Carretero, M. J. Casanove, K. Bouzehouane, S. Guyard, E. Larquet, and J. P.
Contour, Electronic conductivity and structural distortion at the interface between insulators
SrTiO3 and LaAlO3, Phys. Status Solidi A 203(9), 2209 (2006).
13. A. Levy, F. Bi, M. Huang, S. Lu, M. Tomczyk, G. Cheng, P. Irvin, J. Levy, Writing and Low-
Temperature Characterization of Oxide Nanostructures, Journal of Visualized Experiments (in
press, 2014)