You are on page 1of 5

Temperature-dependence of the Hall coefficient of NdNiO3 thin films

Adam J. Hauser, Evgeny Mikheev, Nelson E. Moreno, Tyler A. Cain, Jinwoo Hwang, Jack Y. Zhang, and
Susanne Stemmer
Citation: Applied Physics Letters 103, 182105 (2013); doi: 10.1063/1.4828557
View online: http://dx.doi.org/10.1063/1.4828557
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/103/18?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Correlation between stoichiometry, strain, and metal-insulator transitions of NdNiO3 films
Appl. Phys. Lett. 106, 092104 (2015); 10.1063/1.4914002
DC current induced metal-insulator transition in epitaxial Sm0.6Nd0.4NiO3/LaAlO3 thin film
AIP Advances 4, 057102 (2014); 10.1063/1.4874642
Anisotropic-strain-controlled metal-insulator transition in epitaxial NdNiO3 films grown on orthorhombic NdGaO3
substrates
Appl. Phys. Lett. 103, 172110 (2013); 10.1063/1.4826678
Effect of structural and magnetic exchange coupling on the electronic transport of NdNiO3 films intercalated with
La0.7Sr0.3MnO3 thin layers
Appl. Phys. Lett. 103, 032403 (2013); 10.1063/1.4813490
Electric currents induced step-like resistive jumps and negative differential resistance in thin films of
Nd0.7Sr0.3MnO3
J. Appl. Phys. 111, 07E131 (2012); 10.1063/1.3675998

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
168.131.40.91 On: Mon, 09 Mar 2015 16:32:46

APPLIED PHYSICS LETTERS 103, 182105 (2013)

Temperature-dependence of the Hall coefficient of NdNiO3 thin films


Adam J. Hauser, Evgeny Mikheev, Nelson E. Moreno, Tyler A. Cain, Jinwoo Hwang,
Jack Y. Zhang, and Susanne Stemmer
Materials Department, University of California, Santa Barbara, California 93106-5050, USA

(Received 18 September 2013; accepted 17 October 2013; published online 31 October 2013)
The Hall coefficient of epitaxial NdNiO3 films is evaluated in a wide range of temperatures, from the
metallic into the insulating phase. It is shown that for temperatures for which metallic and insulating
regions co-exist, the Hall coefficient must be corrected for the time-dependence in the longitudinal
resistance, which is due to a slow evolution of metallic and insulating domains. The positive Hall and
negative Seebeck coefficients, respectively, in the metallic phase are characteristic for two bands
participating in the transport. The change in the sign of the Hall coefficient to negative values in the
insulating phase is consistent with the suppression of the contribution from the large hole-like Fermi
C 2013 AIP Publishing LLC.
surface, i.e., the formation of a (pseudo)gap due to charge ordering. V
[http://dx.doi.org/10.1063/1.4828557]
The rare earth nickelates (chemical formula RNiO3,
where R is a trivalent rare earth ion) undergo a metal-to-insulator transition (for R
6 La) upon cooling that has generated
significant interest for understanding charge and spin ordering
phenomena in correlated materials1 and for potential application in novel switching devices.25 NdNiO3 is a prototype
RNiO3, exhibiting a metal-to-insulator transition (MIT) at a
temperature (TMIT) of 200 K (in bulk)6 that is accompanied
by a lowering of the symmetry from orthorhombic to monoclinic, charge ordering,79 and a complex antiferromagnetic
state.1012 The MIT is first-order, resulting in the co-existence
of metallic and insulating regions over a finite temperature
range, and time-dependence of the transport coefficients
within the two-phase region.13,14 The Hall coefficient (RH) is
a useful tool for the study of changes in the electronic structure at the MIT. Unusual, non-monotonic features have been
reported recently in the temperature-dependence of RH of
RNiO3s.5,15 In this letter, we discuss studies of the resistivity,
RH, and the Seebeck coefficient of NdNiO3 films. We show
that for measurements at temperatures that lie within the phase
coexistence region, RH must be corrected for the drift in
the resistivity over time. We interpret the results for the
drift-corrected RH in terms of the current understanding of the
electronic states of NdNiO3.
Epitaxial NdNiO3 films were grown on (001) LaAlO3
by rf magnetron sputtering at a substrate temperature of
700  C. The total growth pressure of a 3:1 Ar:O2 mixture
was 300 mTorr and the rf power was 80 W, yielding a
growth rate of 3.3 nm/h. High-resolution x-ray diffraction
(XRD) measurements (Phillips MRD XPert Thin Film
Diffractometer) and high-angle annular dark-field (HAADF)
scanning transmission electron microscopy (STEM) were
performed to confirm the epitaxial orientation relationships.
In the following, we index all reflections and planes using
pseudo-cubic unit cells for the NdNiO3 films and the rhombohedral substrate, respectively. Films were patterned into
Hall bar structures (300 lm channel width) via contact lithography. Ohmic contacts of Ni(20 nm)/Au(300 nm) were
deposited by electron beam evaporation, and device isolation
was achieved with a wet etch of 25% HCl in water.
Measurements of the in-plane longitudinal resistivity (Rxx),
0003-6951/2013/103(18)/182105/4/$30.00

the Hall resistivity (Rxy), and the Seebeck coefficient (S) as a


function of temperature were performed using a Quantum
Design Physical Properties Measurement System (PPMS).
Seebeck coefficient measurements are reported in the metallic phase only, where the error is less than 1% [this uncertainty rises sharply below TMIT due to the large change in
resistance with temperature, and high resistivity of the insulating phase, respectively]. Rxx was measured between 300 K
and 2 K upon cooling and heating. Rxy and Rxx were measured simultaneously from 130 K to 20 K upon cooling from
room temperature and back upon heating from 10 K, using a
magnetic field (B) sweeps between 69 T. Time-dependent
measurements were carried out for up to several hours, as
described below.
Figure 1(a) shows an on-axis XRD pattern for a
16.5-nm-thick NdNiO3 film. Thickness fringes suggest a

FIG. 1. (a) On-axis XRD around the 002 reflection of a 16.5 nm NdNiO3
film grown on (001) LaAlO3. (b) Off-axis XRD of the same film at a tilt
angle of W 45 , showing 011 reflections of film and substrate. The double
peak of the LaAlO3 substrate is due to twinning. (c) Cross-section
HAADF-STEM image. (d) Longitudinal resistivity of a 16.5 nm film as a
function of temperature measured upon cooling and heating, respectively.

103, 182105-1

C 2013 AIP Publishing LLC


V

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
168.131.40.91 On: Mon, 09 Mar 2015 16:32:46

182105-2

Hauser et al.

smooth surface and agree with the film thickness measured


,
in STEM. The out-of-plane film lattice spacing is 3.836 A
larger than that of bulk NdNiO3 (pseudo-cubic lattice param (Ref. 16)), due to the compressive in-plane
eter 3.81 A
strain imposed by the LaAlO3 (pseudo-cubic lattice parame ). Off-axis XRD (W 45 with respect to the surter 3.79 A
face normal) around the 011 reflection also shows thickness
fringes [Fig. 1(b)]. The film peak position corresponds to a
and an in-plane lattice spacing
lattice spacing d011 of 2.697 A

of 3.792 A, identical to that of the substrate. The films are


therefore coherently strained to the substrate. Figure 1(c)
shows a HAADF-STEM image viewed along [010], confirming epitaxial growth. Figure 1(d) shows that the films exhibit
a MIT at 100 K, with a hysteresis that has a width of
25 K. TMIT is comparable to that of NdNiO3 thin films
reported in the literature,3,17 but is lower than that of bulk,
which typical for thin RNiO3 films.
Figure 2(a) shows Rxx as a function of time at different
temperatures around the MIT. Within the temperature range
of the hysteresis seen in Fig. 1(d), significant time-dependence
is apparent in Rxx that persists to longer than 810 hr, i.e., the
maximum observation period in this study. Figure 2(b) shows
a measure of the magnitude of the drift as a function of temperature (T), which correlates with the hysteresis in Rxx(T) and
is roughly proportional to the slope of Rxx(T). This time dependence has been reported previously for both bulk and thin
film NdNiO3,14,18 where it was described by a stretched
exponential. It reflects the time scales associated with the formation of metallic (heating above TMIT) or insulating phases
(cooling below TMIT), respectively. A single, simple exponential could not describe our data, but both a stretched exponential decay or the sum of two exponential decay functions with
different time constants (103 and 104 s) provided an excellent fit.
In Hall measurements a small parasitic contribution due
to Rxx exists due to the finite contact lead widths. This results
in a non-linear Rxy as a function of B, as well as a vertical
offset in Rxy B. These are commonly corrected with a vertical offset and removal of the part that is symmetric about
B 0. For a non-ferromagnetic material, however, Rxy B
will still trace back over itself closely upon reversal of B.
Figure 2(c) shows the effect of the drift on consecutive
Rxy B traces recorded at 80 K, i.e., within the hysteretic regime (dashed lines). Here, B was swept back and forth form
the two maximum fields for many hours. Subtracting the
modeled longitudinal resistance Rxx and multiplying by a
scaling factor yields a time-corrected Hall measurement
[shown as solid lines in Fig. 2(c)] that yields a RH similar to
the one that the uncorrected data is converging to. We note
that averaging one or several sweeps, even when recorded
over the large time interval of 6 h shown in Fig. 2(c), does
not give correct values for RH, due to the long time constants
involved.19
Figures 3(a) and 3(b) show the resistivity and driftcorrected RH as a function of temperature around the MIT.
Like the longitudinal resistivity, RH exhibits hysteretic
behavior between 50 and 120 K. In the metallic region,
RH is positive, as previously reported for NdNiO3 (Refs. 3
and 20) and other RNiO3s.21,22 Transport in the metallic
RNiO3s is generally understood to be determined by two

Appl. Phys. Lett. 103, 182105 (2013)

FIG. 2. (a) Time-dependence of the longitudinal resistance at different temperatures around TMIT. Shown is the percentage change from the value measured 5 min after the temperature has stabilized, upon cooling from room
temperature (filled symbols) and heating from 10 K (open symbols), respectively. (b) Amount of drift, as defined in (a), after 5 h, as a function of temperature under cooling (blue filled circles) and heating (orange open triangles),
respectively. (c) Raw Rxy data (dashed lines) as a function a magnetic field
measured over 6 h of continuous field sweeps from 9 to 9 T at T 80 K.
Also shown is the drift-corrected Rxy (solid line).

bands that cross the Fermi level, giving rise to a small electron pocket and a large hole Fermi surface, respectively.2326
The large hole Fermi surface dominates RH. In contrast, the
Seebeck coefficient, shown in Fig. 3(c) is negative,

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
168.131.40.91 On: Mon, 09 Mar 2015 16:32:46

182105-3

Hauser et al.

Appl. Phys. Lett. 103, 182105 (2013)

resistance in the two-phase region. We find a positive and


temperature-independent RH in the metallic region that is
dominated by the large hole Fermi surface despite the
presence of a small electron pocket. RH becomes
temperature-dependent and negative below the MIT, which
is consistent with the hole Fermi surface opening up a
(pseudo)gap, causing a decrease in the hole concentration,
and RH becoming dominated by residual electron carriers.
The authors thank Leon Balents and Jim Allen for helpful discussions. This work was supported in part by FAME,
one of six centers of STARnet, a Semiconductor Research
Corporation program sponsored by MARCO and DARPA.
A.J.H. acknowledges support through an Elings Prize
Fellowship of the California Nanosystems Institute at
University of California, Santa Barbara. The work made use
of central facilities of the UCSB MRL, which is supported
by the MRSEC Program of the National Science Foundation
under Award No. DMR-1121053. The work also made use
of the UCSB Nanofabrication Facility, a part of the NSFfunded NNIN network.
1

M. L. Medarde, J. Phys.: Condens. Matter 9, 1679 (1997).


J. Son, S. Rajan, S. Stemmer, and S. J. Allen, J. Appl. Phys. 110, 084503
(2011).
3
R. Scherwitzl, P. Zubko, I. G. Lezama, S. Ono, A. F. Morpurgo, G.
Catalan, and J.-M. Triscone, Adv. Mater. 22, 5517 (2010).
4
S. Asanuma, P.-H. Xiang, H. Yamada, H. Sato, I. H. Inoue, H. Akoh, A.
Sawa, K. Ueno, H. Shimotani, H. Yuan et al., Appl. Phys. Lett. 97,
142110 (2010).
5
W. L. Lim, E. J. Moon, J. W. Freeland, D. J. Meyers, M. Kareev, J.
Chakhalian, and S. Urazhdin, Appl. Phys. Lett. 101, 143111 (2012).
6
J. S. Zhou, J. B. Goodenough, and B. Dabrowski, Phys. Rev. Lett. 94,
226602 (2005).
7
U. Staub, G. I. Meijer, F. Fauth, R. Allenspach, J. G. Bednorz, J.
Karpinski, S. M. Kazakov, L. Paolasini, and F. dAcapito, Phys. Rev. Lett.
88, 126402 (2002).
8
J. L. Garca-Mu~
noz, M. A. G. Aranda, J. A. Alonso, and M. J. MartinezLope, Phys. Rev. B 79, 134432 (2009).
9
I. I. Mazin, D. I. Khomskii, R. Lengsdorf, J. A. Alonso, W. G. Marshall,
R. A. Ibberson, A. Podlesnyak, M. J. Martinez-Lope, and M. M. AbdElmeguid, Phys. Rev. Lett. 98, 176406 (2007).
10
V. Scagnoli, U. Staub, A. M. Mulders, M. Janousch, G. I. Meijer, G.
Hammerl, J. M. Tonnerre, and N. Stojic, Phys. Rev. B 73, 100409(R)
(2006).
11
S. Lee, R. Chen, and L. Balents, Phys. Rev. Lett. 106, 016405 (2011).
12
B. Lau and A. J. Millis, Phys. Rev. Lett. 110, 126404 (2013).
13
D. Kumar, K. P. Rajeev, J. A. Alonso, and M. J. Martinez-Lope, J. Phys.:
Condes. Matter 21, 485402 (2009).
14
D. Kumar, K. P. Rajeev, A. K. Kushwaha, and R. C. Budhani, J. Appl.
Phys. 108, 063503 (2010).
15
S. D. Ha, R. Jaramillo, D. M. Silevitch, F. Schoofs, K. Kerman, J. D.
Baniecki, and S. Ramanathan, Phys. Rev. B 87, 125150 (2013).
16
J. L. Garca-Mu~
noz, J. Rodriguez-Carvajal, P. Lacorre, and J. B. Torrance,
Phys. Rev. B 46, 4414 (1992).
17
G. Catalan, R. M. Bowman, and J. M. Gregg, J. Appl. Phys. 87, 606
(2000).
18
D. Kumar, K. P. Rajeev, J. A. Alonso, and M. J. Martinez-Lope, Phys.
Rev. B 88, 014410 (2013).
19
See supplementary material at http://dx.doi.org/10.1063/1.4828557 for
results of RH obtained by time-averaging plots of sweeps of Rxy as a function of B at different temperatures.
20
J. Son, B. Jalan, A. P. Kajdos, L. Balents, S. J. Allen, and S. Stemmer,
Appl. Phys. Lett. 99, 192107 (2011).
21
S. W. Cheong, H. Y. Hwang, B. Batlogg, A. S. Cooper, and P. C. Canfield,
Physica B 194196, 1087 (1994).
22
J. Son, P. Moetakef, J. M. LeBeau, D. Ouellette, L. Balents, S. J. Allen,
and S. Stemmer, Appl. Phys. Lett. 96, 062114 (2010).
23
N. Hamada, J. Phys. Chem. Solids 54, 1157 (1993).
2

FIG. 3. (a) Longitudinal resistivity q and drift-corrected RH as a function of


temperature. (b) Same as (a) but within a narrower range of temperatures to
show the positive RH in the metallic region. (c) Seebeck coefficient as a
function of temperature in the metallic phase. Measurements upon cooling
(heating) are shown as circles (triangles).

consistent with previous reports for NdNiO3 and related


RNiO3.15,22,27,28 The difference in the signs of the two transport coefficients confirms the two-band nature of the electrical transport in metallic NdNiO3. Furthermore, RH is nearly
temperature independent in the metallic region but becomes
temperature dependent below TMIT. In the insulating regime,
RH is negative and changes linearly with temperature. Down
to the lowest temperatures, RH does not show any unusual
non-monotonic behavior. The sign change near TMIT is consistent with a (pseudo)gap opening in the large hole Fermi
surface and/or its gradual disappearance, causing the hole
density to decrease with temperature. This will cause the
electron pocket to dominate RH and result in the observed
temperature dependence. The formation of a (pseudo)gap in
the insulating phase is consistent with charge ordering,
which has been well-established both theoretically and
experimentally for the insulating phase of NdNiO3.710,12,29
In summary, we have shown that the correct evaluation
of the Hall effect of RNiO3 films that undergo a MIT
requires correcting for the drift of the longitudinal

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
168.131.40.91 On: Mon, 09 Mar 2015 16:32:46

182105-4
24

Hauser et al.

R. Eguchi, A. Chainani, M. Taguchi, M. Matsunami, Y. Ishida, K. Horiba,


Y. Senba, H. Ohashi, and S. Shin, Phys. Rev. B 79, 115122 (2009).
25
S. B. Lee, R. Chen, and L. Balents, Phys. Rev. B 84, 165119 (2011).
26
H. K. Yoo, S. I. Hyun, L. Moreschini, Y. J. Chang, D. W. Jeong, C. H.
Sohn, Y. S. Kim, H.-D. Kim, A. Bostwick, E. Rotenberg et al., e-print
arXiv:1309.0710[cond-mat.str-el].

Appl. Phys. Lett. 103, 182105 (2013)


27

X. Q. Xu, J. L. Peng, Z. Y. Li, H. L. Ju, and R. L. Greene, Phys. Rev. B


48, 1112 (1993).
K. P. Rajeev, G. V. Shivashankar, and A. K. Raychaudhuri, Solid State
Commun. 79, 591 (1991).
29
T. Mizokawa, D. I. Khomskii, and G. A. Sawatzky, Phys. Rev. B 61,
11263 (2000).
28

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
168.131.40.91 On: Mon, 09 Mar 2015 16:32:46