Journal of Physics: Condensed Matter

PAPER You may also like

- Soft x-ray spectroscopic investigation of
Switching of majority charge carriers by Zn doping Zn doped CuCl produced by pulsed dc
magnetron sputtering
in NdNiO3 thin films K V Rajani, S Daniels, P J McNally et al.

- Characterization of Zn doped SnO2 thin

films prepared by the SILAR technique for
optoelectronic applications
To cite this article: Kavita Soni et al 2021 J. Phys.: Condens. Matter 33 015602
Muhammed Emin GÜLDÜREN

- Multifold polar states in Zn-doped

Sr0.9Ba0.1TiO3 ceramics
Yan-Yan Guo, , Yun-Jun Guo et al.
View the article online for updates and enhancements.

This content was downloaded from IP address 143.248.16.133 on 25/05/2023 at 03:43

 Journal of Physics: Condensed Matter

J. Phys.: Condens. Matter 33 (2021) 015602 (8pp) https://doi.org/10.1088/1361-648X/abb864

Switching of majority charge carriers

by Zn doping in NdNiO3 thin films
Kavita Soni1 , Harisankar S1 , Mahesh Chandra1,3 , Parasmani Rajput2 and
K R Mavani1,∗
1
Discipline of Physics, Indian Institute of Technology (IIT) Indore, Khandwa Road, Simrol, 453 552,
India
2
Atomic & Molecular Physics Division, Bhabha Atomic Research Center (BARC), Trombay, Mumbai
400 085, India

E-mail: [email protected]

Received 31 July 2020

Accepted for publication 14 September 2020
Published 13 October 2020

Abstract
We have studied the effects of Zn doping on the structural and electronic properties of
epitaxial NdNiO3 thin films grown on single-crystal LaAlO3 (001) (LAO) substrates by pulsed
laser deposition. The films are deposited in two sets, one with variation in Zn doping, and
another with variation in thickness for undoped and 2% Zn doping. The experimental
investigations show that Zn occupies Ni-site and that the films are grown with an in-plane
compressive strain on LAO. All the films show metal-to-insulator transitions with a
thermal hysteresis in the temperature-dependent resistivity curves except 5% Zn-doped film,
which remains metallic. The theoretical fits show non-Fermi liquid behaviour, which gets
influenced by Zn doping. The Hall resistance measurements clearly show that Zn doping
causes injection of holes in the system which affects the electronic properties as follows:
i) the metallic conduction increases by two factors just by 0.5% Zn doping whereas, 5%
doping completely suppresses the insulating state, ii) a reversal of the sign of Hall coefficient
of resistance is observed at low temperature.
Keywords: perovskite RNiO3 , thin films, metal-to-insulator transition, correlated electron
system, Zn-doped NdNiO3
(Some figures may appear in colour only in the online journal)

1. Introduction phenomena [4], orbital engineering and quantum-criticality

related features make them good candidates for ther-
Metal-to-insulator (MI) transitions in RNiO3 (R = rare earth mochromic coating, atmospheric sensors [2], photovoltaic
ions) materials have many technological applications such cells [5], non-volatile memories, electric-field sensors [6] etc.
as in thermal sensors and switches. Due to highly tunable There are also some theoretical predictions of multiferroicity
properties of RNiO3 thin films, this family of oxides pro- [7], superconductivity [8] and topological insulating phases
vide an excellent opportunity to explore not only funda- [9] which open the door for new investigations. For these rea-
mental properties but also the control on the functions from sons, RNiO3 nickelates have attracted much attention of the
the applications point of view. These materials show ther- scientific community working on oxides.
mochromic effect [1], reversible change in resistance [2], elec- The perovskite-type RNiO3 family of oxides has a rich
tric field controlled property modulation [3], charge-ordering phase diagram [10, 11]. In this structure, the rare-earth cation
∗
located at the centre of a unit cell and another metal cation
Address for correspondence. Krushna R. Mavani (PhD) Professor, Discipline
of Physics Indian Institute of Technology Indore.
Ni sits at the corners of NiO6 octahedra. As R changes from
3
Present address. Department of Physics, Indian Institute of Technology La to Pr, Nd, Sm and so on with decreasing ionic radii, the
Delhi, Hauz Khaz, New Delhi 110016, India. NiO6 octahedra undergo distortion in order to support the

1361-648X/20/015602+8$33.00 1 © 2020 IOP Publishing Ltd Printed in the UK

J. Phys.: Condens. Matter 33 (2021) 015602 K Soni et al

stable structure. RNiO3 (R = La) shows the first-order MI tran- targets were prepared by solid-state reaction method by mix-
sition at low temperatures, whereas LaNiO3 remains metallic ing high purity powders of Nd2 O3 , NiO and ZnO in appropriate
throughout the temperature range. The MI transition tempera- molar ratios and subsequent heating treatments with intermit-
ture (T MI ) shifts towards the higher temperature with decreas- tent grinding. The excimer KrF laser of wavelength 248 nm
ing R cation size because of the change in Ni–O–Ni bond angle was used for ablating the target material. For all the deposi-
and bond lengths. This bending, breathing and straightening tions, the laser energy was set to 310 mJ, and the pulse repe-
in bond angles and bond lengths are predominantly responsi- tition rate was kept at 4 Hz. Single crystal LaAlO3 (001) was
ble for defining the overlapping of Ni 3d and O 2p orbitals, used as a substrate and was cleaned sequentially in acetone and
and hence the electronic properties. Among all the bulk nicke- propanol prior to deposition. Target to substrate distance was
lates, only PrNiO3 and NdNiO3 exhibit simultaneous occur- fixed at 4.5 cm, and the substrate temperature was maintained
rence of electronic and magnetic phase transitions [12–15]. at 993 K during deposition. The substrate was slowly cooled
According to Zaanen Sawatzky–Allen framework, RNiO3 lies at the rate of 3 K min−1 till 673 K followed by 5 K min−1
in the category of charge transfer insulators where Coulomb rate till ambient temperature. Before ablation, the chamber was
repulsion energy U which is the energy difference between evacuated, and then the pure oxygen gas was infused into the
unoccupied and occupied 3d bands, is more than the energy chamber and maintained at a partial pressure of 40 Pa for the
difference between the anion 2p band and unoccupied 3d band deposition. After the deposition, the films were annealed in
(Δ), also known as charge transfer gap [16]. The Fermi level partial oxygen pressure at 1000 Pa for 3 min at 993 K. By
lies in this gap at the low-temperature regime and thus insu- using the same parameters, we have prepared two sets of films.
lating state become apparent, but at the higher temperature In set-1, the films were grown at a fixed thickness of 10 nm
side the thermal broadening of bands take place and ultimately with variation in doping concentration (x = 0, 0.005, 0.02,
charge-transfer gap tends to zero and metallicity appears [13]. 0.05), and in set-2, the films were grown at a fixed thickness of
The origin of T MI is also discussed in the framework of charge 30 nm for x = 0 and x = 0.02.
disproportionation or charge ordering at nickel site [4, 17]. The phase purity of thin films and the structural orientations
A systematic tuning of T MI has been observed by varying were examined by x-ray diffraction (XRD) measurements. To
many different parameters in thin films, such as by changing study the in-plane lattice of films, the reciprocal space map
the dimensionality (bulk to 2D and 1D) [18], doping [19–23], (RSM) was recorded using x-ray diffractometer (Emperean,
varying oxygen content [24–26], application of electric field PanAlytical) with five-axes cradle (Cu Kα). Room tempera-
or voltage [27, 28], strain and thickness variations [18, 29, 30]. ture Ni K-edge x-ray absorption near edge structure (XANES)
One of the most controlling methods to tune electronic proper- spectra of 10 nm thin films were recorded in the fluorescence
ties is by injecting charge carriers via doping. The doping can mode at the BL-09, scanning extended x-rays absorption fine
be done at R site as well as Ni site. In bulk NdNiO3 , injection structure beamline of INDUS-2 synchrotron source, RRCAT,
of divalent Sr2+ (hole doping) and tetravalent Th4+ (electron indore using vortex energy-dispersive detector. The beam-
doping) are responsible for lowering the T MI , and only for the line consists of Rh/Pt coated meridional cylindrical mirror for
5% of divalent Sr2+ leads the complete suppression of insu- collimation and Si (111) based double crystal monochroma-
lating nature [22, 31], indicating that hole doping is more effi- tor to select excitation energy. The energy range of XAFS
cient in bringing metallicity. Rather than doping at R site, 3d was calibrated using Ni metal foil at 8333 eV. The XANES
metal ion doping at Ni-site can create a very precise and direct spectra normalisation were done using ATHENA software.
impact on the electronic bands. Earlier electron doping at Ni- The temperature-dependent resistivity measurements were
site by hydrogenation in SmNiO3 thin films was reported to performed using the four-probe method in a Janis-made
induce switching of resistance and band-gap modulation [32]. closed-cycle cryostat (300 K to 8 K). Physical properties mea-
Here, hydrogenation can alter the two resistivity states in the surement system of quantum design was used with Van-der
same film. Also, the transition metals like Cu, Fe, Co, are used Pauw geometry for measuring Hall resistance of the films
as the dopant in bulk LaNiO3 and NdNiO3 [33–35]. However, with the varying magnetic field up to 7 T at three different
in thin films, there are only a few studies performed on Ni- temperatures of 30 K, 100 K and 300 K.
site doping [19, 20, 23, 36]. In our earlier investigations, the
doping of electrons via larger size tetravalent Mn at Ni site 3. Results and discussions
under strain (compressive and tensile) was found responsible
for shifting the T MI at higher temperatures and bring a com- Figure 1 shows the full range XRD patterns of NdNi1−x Znx O3 ,
plete insulating state [19, 23]. With the above points in mind, (x = 0, 0.005, 0.02, 0.05) thin films of set 1 and set 2. There are
we have studied the effects of hole injection via Zn doping in no impurities peaks observed. The XRD patterns around (002)
NdNiO3 thin and thick films and observed several systematic Bragg peak and RSMs of all the films around (301) Bragg
variations in the electronic properties. peak are shown in figure 2 (for set-1) and figure 3 (for set-
2). The films grew with high crystallinity under a compressive
2. Experimental strain on the LAO substrate. The pseudocubic lattice parame-
ter of bulk NdNiO3 is 3.80 Å, whereas that of LAO (100) is
Two sets of NdNi1−x Znx O3 , (x = 0, 0.005, 0.02, 0.05) thin 3.79 Å. Thus the mismatch of lattice constant in thin films is
films (set-1: 10 nm and set-2: 30 nm) were grown by pulsed nearly −0.26%. To investigate the in-plane strain, RSM was
laser deposition system using a 20 mm diameter target. The performed around the asymmetric (301) Bragg reflection. It

2
J. Phys.: Condens. Matter 33 (2021) 015602 K Soni et al

Figure 3. RSM and XRD plots of NdNi1−x Znx O3 (x = 0, 0.02) thin

films (30 nm) grown on LAO (001) substrate. (Here rlu abbreviated
as reciprocal lattice unit).

Figure 1. XRD patterns of NdNi1−x Znx O3 (x = 0, 0.005, 0.02, 0.05)

thin films (10 nm, 30 nm) grown on LAO (001) substrate.

Figure 2. RSM and XRD plots of NdNi1−x Znx O3 (x = 0, 0.005,

Figure 4. Normalized absorption XANES spectra at Ni K-edge of
0.02, 0.05) thin films (10 nm) grown on LAO (001) substrate. (Here
Ni metal foil, NiO standard and NdNi1−x Znx O3 (x = 0, 0.005, 0.02,
rlu abbreviated as reciprocal lattice unit).
0.05) thin films. The inset figure shows the first derivative of the
normalised absorption coefficient versus energy curves. (Here μ(E)
is an absorption coefficient).
can be seen in the given figures that the Bragg reflection of
the substrate and thin films (x = 0, 0.005 and 0.02) lie on the
same pseudomorphic line. With these results, we can conclude present case, Ni+4 conversion does not maintain such stability
that the films of set-1 are grown epitaxial except for x = 0.05, at the higher percentage of Zn doping, and the pellets broke
whereas the films of set-2 are slightly relaxed. In figure 2, the into pieces soon after sintering. Figure 3 shows the XRD plots
(002) XRD peaks of thin films from set-1 systematically shift- of x = 0 and x = 0.02 thin films of 30 nm thickness. A split-
ing towards the higher 2θ value with increasing Zn doping ting of x = 0.02 film’s peak is clearly visible in the XRD plot
up to x = 0.02. This shows that Zn doping creates a signifi- and RSM also, which was absent for 10 nm film. Such split-
cant impact on out of plane lattice parameter. The divalent Zn ting indicates a structural transformation to plausibly a lower
ion decreases the out of plane lattice parameter which causes symmetry or a formation of a thin (∼4–5 nm) dead-layer at the
a reduction in the unit cell volume with increased doping bottom. Such dead-layer formation has been earlier observed
amount. Thus, this systematic change in volume suggests that for perovskite manganite thin films [37].
the divalent doping of Zn at Ni site leads to a change of valence To confirm the oxidation state of Ni in NdNi1−x Znx O3
state of Ni from larger size Ni+3 (r = 0.56 Å) to smaller size (x = 0, 0.005, 0.02, 0.05) thin films, the experiment was per-
Ni+4 (r = 0.48 Å) ion in order to maintain the charge neutral- formed with NiO reference sample. Figure 4 shows the nor-
ity. A similar observation was earlier made for NdNi1−x Mnx O3 malised XANES spectra versus energy curve at the Ni K-edge.
series of films where the unit cell volume expanded by Mn+4 It is clearly observed that all the thin films show oxidation
doping at Ni-site because of Ni+3 to Ni+2 conversion [23]. The state higher than 2+ of NiO reference. It is important to note
divalent state of Ni is more stable, Mn-doping up to higher per- that there is a decrease in white-line intensity, i.e. an intense
centage maintained the stability of a structure. However, in the absorption in the near-edge as a function of increasing Zn

3
J. Phys.: Condens. Matter 33 (2021) 015602 K Soni et al

complete metallicity [14]. Also, the compressive strain favours

the metallic state in RNiO3 compounds [29, 40]. The combi-
nation of these two effects drive T MI at the lower temperature,
and the overall resistivity decreases in both the metallic as well
as the insulating region with increased doping up to x = 0.02.
Minimum resistivity is observed for x = 0.02 thin film at all
the temperatures. Although the film with x = 0.05 has metallic
nature throughout the temperature range, it shows the highest
resistivity. The plausible reason may be an incoherent distri-
bution of compressive strain along the thickness (figure 2) and
aggravation in charge carriers scattering. On the other hand,
if we compare the resistivity of set-1 and set-2 films, it is
clear that the set-2 films exhibit high resistivity along with
somewhat overlapped values as compared to the films of set-1.
The reason for this behaviour is that the set-1 films are epi-
taxial while the set-2 films are relaxed as confirmed from
RSM. For investigating the effects of Zn doping and com-
pressive strain on the metallic nature of these films, we have
calculated the temperature coefficient of resistivity (α) around
300 K. The change of α with respect to doping concentration x
is shown in table 1. It is observed that α increases initially up to
x = 0.02 and then suddenly it decreases for x = 0.05 due to the
difference in the ground state.
The thicker films are slightly relaxed as compared to the
Figure 5. Temperature dependent resistivity (ρ) of NdNi1−x Znx O3 thin films. C Girardot et al have reported that, when the thick-
(x = 0, 0.005, 0.02, 0.05) thin films (a) 10 nm, (b) 30 nm, grown on
LAO (001) substrate.
ness of NdNiO3 thin film decreases, it suffers more influence
from the template effect of LAO substrate and the Ni–O–Ni
bond is straightened [41, 42]. Therefore, more distortion
doping, which confirms that the occupancy of Zn is observed occurs in NiO6 octahedra for thicker films, which increase T MI
at Ni site only. At 8332.3 eV a pre-edge is observed due to due to the thickness (table 2). T MI is determined by the dρ/dT
dipolar transition between metal 1s states and hybridised metal versus T plots. In table 2, the transition hysteresis represents
3d–O 2p bands [38, 39]. The inset of figure 4 shows the 1st the difference in T MI of heating and cooling processes, which
derivative of XANES spectra of undoped and doped NdNiO3 increases with Zn doping, indicating an effect of disorder of
thin films and NiO2+ standard that provides a clear vision the system.
about the valance state of Ni in all the thin films. For analysing the governing conduction mechanism in
Figure 5(a) shows the temperature-dependent resistivity of Zn-doped nickelates thin films, the metallic region of the
set-1 thin films (10 nm). The T MI systematically decreases as temperature-dependent resistivity curves are fitted with the
the doping increases until x = 0.02. The obtained value of T MI power-law equation,
for x = 0, 0.005, 0.02 are respectively at 120 K, 115 K, 105 K
while heating, and 115 K, 105 K, 90 K while cooling. It can ρNFL (T) = ρ0 + ρn T n (1)
be deduced that the T MI possesses an inverse linear relation-
where, n is an integer or non-integer value (n < 2 here, n =
ship (for both thick and thin films) with Zn doping. However,
4/3 for non-Fermi liquid (NFL) behaviours), n = 1 relates
this relation is more pronounced for the 10 nm thin films. It is
to the scattering of charge carriers due to phonon vibra-
also seen that the hysteresis has nearly linear dependence to the
tions, ρ0 known as the residual resistivity which is inde-
doping percent for 10 nm thin films which are grown epitaxi-
pendent of temperature, ρ1 is known as strength of elec-
ally. We observed the complete different resistivity behaviour
tron–phonon scattering, and n = 2 corresponds to Fermi liquid
of film x = 0.05, which shows metallicity throughout the tem-
behaviour. Here, the linear relation between ρ(T ) and T (n
perature range. Also for set-2 (figure 5(b)), the same feature
= 1) fails to describe the experimental data, and the ongo-
is observed; even if the thickness is different, the transition
ing recent studies revealed the NFL behaviour could exist
temperature is shifting towards the lower temperature.
in the RNiO3 system; therefore we also take into account
Hall resistivity measurements were performed for undoped
the resistivity saturation (ρSAT ) by following this equation
and doped (x = 0.02, 0.05) thin films (10 nm) and explained in
[18, 43],
detail in the subsequent section. It shows that Zn doping injects
ρ−1 (T) = ρ−1 −1
NFL (T) + ρSAT . (2)
holes in the 3d band and therefore, there is an enhancement in
the density of charge carriers in the system. Thus, the broad- Here ρSAT act as a parallel resistor, which shows resis-
ening of electronic bands may come into the picture. Due to tivity saturation at high temperature. The set-1 films show a
this broadening of the bands, the overlapping of oxygen 2p and good fit to equation (2) in the temperature range 300–120 K,
Ni 3d orbitals increases which finally leads the system towards except for x = 0.05. After many trials, this particular with

4
J. Phys.: Condens. Matter 33 (2021) 015602 K Soni et al

Table 1. Parameters derived from temperature-dependent resistivity data of NdNi1−x Znx O3 (x = 0, 0.005, 0.02, 0.05) thin films grown on
LAO (001) substrate.
Transition sharpness Temperature coefficient
x t (nm) T MI (K) cooling T MI (K) heating Hysteresis (K) −d(ln ρ)/dT (Ω cm K−1 ) of resistivity α (K−1 )
0 10 115 120 5 7.02 × 10−3 0.526(0)
0.005 10 105 115 10 1.15 × 10−2 0.863(7)
0.02 10 90 105 15 3.91 × 10−3 1.160(1)
0.05 10 No transition — — — 0.407(5)
0 30 170 155 25 9.42 × 10−3 0.129(4)
0.02 30 140 120 20 2.31 × 10−2 0.433(7)

Table 2. Fitting parameters of temperature-dependent resistivity curves of NdNi1−x Znx O3 (x = 0, 0.005, 0.02, 0.05) thin films grown on
LAO (001) substrate.
x t (nm) n Temp. range ρ(300K) (Ω cm) Residual resistivity (Ω cm) A (Ω cm (K)−n ) ρSAT (Ω cm)
0 10 4/3 125–300 K 1.06 × 10−3 7.88 × 10−4 3.14 × 10−7 (578 ± 1.90) × 10−5
0.005 10 4/3 115–300 K 9.51 × 10−4 5.70 × 10−4 3.60 × 10−7 (455 ± 1.78) × 10−5
0.02 10 4/3 115–300 K 5.51 × 10−4 4.80 × 10−4 3.90 × 10−7 (108 ± 0.085) × 10−5
0.05 10 4/3 90–300 K 2.07 × 10−3 3.19 × 10−5 7.99 × 10−6 (246 ± 0.390) × 10−5
0.05 10 2 8–90 K 2.07 × 10−3 1.16 × 10−3 2.68 × 10−8 —
0 30 4/3 175–300 K 2.46 × 10−3 2.03 × 10−3 2.60 × 10−7 —
0.02 30 4/3 125–300 K 2.79 × 10−3 1.19 × 10−3 9.50 × 10−7 —

Figure 6. Hall resistivity (ρH ) versus magnetic field (B) plots of NdNi1−x Znx O3 (x = 0, 0.02, 0.05) thin films grown on LAO (001) substrate
at different temperatures.

5% Zn doping, with a relaxed state of strain and complete shown in table 2, indicating that Zn doping is highly influenc-
metallicity, shows a fit to Fermi liquid model [equation (1)] ing in tuning the electronic properties of these films.
with n = 2 from temperature 90–8 K, whereas it shows NFL In order to get the sign of charge carriers involved in con-
state above 90 K (figure 5). It is important to note that this duction mechanism, Hall resistivity measurements were per-
film shows a transformation behaviour from FL-like to NFL- formed in Van der Pauw geometry with the magnetic field up
to 7 T. Figure 6 shows the Hall resistivity versus magnetic
like close to the temperature coinciding with T MI of the other
field plots for three 10 nm thin films of x = 0, x = 0.02 and
films. Also, this film clearly shows a saturating resistivity cur- x = 0.05. The undoped and Zn-doped (x = 0.02, 0.05) films
vature near room temperature, and a similar value of ρSAT is show that the majority charge carriers are holes at room tem-
achieved by the fit too (table 2), which validates the theoreti- perature. For the undoped film, a switching of majority charge
cal fits. Both the films of the set-2 show fit to equation (2) with carriers is observed from hole-like to electron-like below
n = 4/3. The fitting parameters with systematic changes are 100 K. This kind of temperature-driven switching has also

5
J. Phys.: Condens. Matter 33 (2021) 015602 K Soni et al

Table 3. Parameters derived from temperature-dependent Hall resistivity data of NdNi1−x Znx O3 (x = 0, 0.02, 0.05)
10 nm thin films grown on LAO (001) substrate. (ep = epitaxial strain, rx = partially relaxed strain).
x T RH × 10−5 (cc/C) μ (cm2 /V-s) p × 1022 (cc−1 ) n × 1022 (cc−1 )
0(ep) 30 −7.57 Insulating state
100 −3.04 Insulating state
300 0.38 0.004 0.89 0.87
0.02(ep) 30 −4.11 Insulating state
100 Hall resistivity parallel to the x-axis
300 9.60 0.174 1.14 0.65
0.05(rx) 30 11.76 0.099 1.24 0.60
100 10.51 0.073 1.21 0.64
300 2.75 0.013 1.00 0.85

been previously reported [24, 44]. For the film x = 0.02, a very this trend as it is having a partially relaxed state of compressive
systematic change in the switching of majority charge carriers strain. It is also noted that lower mobility of charge carriers is
is observed. At room temperature, the majority charge carri- observed for the present films than reported earlier [44–47].
ers are holes, but at 100 K [T MI (cooling) = 90 K  100 K  The present films are very thin (10 nm) and strained as com-
T MI (heating) = 105 K] there is not much variation in Hall pared to the reported ones (∼60 nm), and this lower mobility
resistivity with respect to the magnetic field. This indicates indicates that the thinness of present films restricts the mobility
that both electrons and holes are almost equally taking part of charge carriers in films.
in conduction. At the sufficiently low temperature of 30 K, the
majority charge carriers are electrons like the undoped film. 4. Conclusions
However, it is important to note for the film x = 0.05 that the
majority charge carriers are holes till 30 K without any such In summary, the structural and electronic properties are stud-
change. Thus, Zn-doped film with complete metallicity does ied for NdNi1−x Znx O3 (0  x  0.02) thin films (10 nm and
not show the change of sign of Hall resistivity. However, the 30 nm) deposited under compressive strain on signal crystal
theoretical fits suggest a transformation of charge conduction LAO (100) substrate. The set-1 films show epitaxial growth,
from NFL-type to Fermi liquid as explained earlier. For this except for x = 0.05, whereas both the films in set-2 with
particular film, the saturating resistivity curve at higher tem- x = 0 and 0.02 exhibit a relaxed state of strain. The 3d ion
peratures is also clearly visible in the plots. Moreover, the non- doping changes the unit cell volume. XANES spectra sug-
linearity observed in Hall resistance plots for x = 0 (figure 6) gested that oxidation state of Ni in the doped and undoped film
does vanish for Zn-doped (x = 0.05) thin film due to the is higher than +2 and the systematic decrease in white-line
absence of a change of sign. The Hall coefficient of resistivity intensity suggested increased doping of Zn at Ni site. The Zn
[RH = dρH /dB] is given in table 3. Thus, we have temperature- doping also caused a tuning of T MI and thermal hysteresis of
driven switching in undoped film and doping induced switch- resistivity at T MI . The increase in thickness caused an increase
ing back to hole like majority charge carriers at low of T MI . Below 100 K, the Hall resistance shows electron-like
temperature. majority charge carriers for x = 0 and x = 0.02 thin film (10
It is well known that nickelates are a multiband system, so nm), which switches to hole-like majority charge carriers for
both electrons and holes contribute to the charge transport phe- higher Zn doping quite systematically. Thus, Zn doping proves
nomenon [24, 45]. As a result, hole and electron carrier densi- to be very efficient in tuning the electronic properties of RNiO3
ties have to be separately calculated from the Hall coefficient films.
(RH ). The Hall coefficient and carrier densities in multiband
systems are related as follows: Acknowledgments
−nμ2n + pμ2p This work is supported by BRNS, India, through a project
RH = (3)
e(nμn + pμ p)2 [37(3)/14/28/2017-BNRS/37225] of KRM.
where n and p are the electron and hole density, respectively,
and μ is their respective mobility. We have taken certain ORCID iDs
assumptions of μp = μn = μ and p + n = 1e− /Ni (i.e. 1 electron
K R Mavani https://orcid.org/0000-0001-7962-6097
per Ni site) to extract the electron (n) and hole carrier densi-
ties (p) and mobility (μ = RH /ρ), as shown in table 3. These References
assumptions, however, are valid only in the metallic regime
[1] Boileau A, Capon F, Laffez P, Barrat S, Endrino J L, Galindo
and are consistent only for RH values less than the order of R E, Horwat D and Pierson J F 2014 Mechanisms of oxida-
10−3 [45–47]. From table 3, it is clear that room temperature tion of NdNiO3−δ thermochromic thin films synthesized by
mobility and the hole concentration is maximum for 2% Zn- a two-step method in soft conditions J. Phys. Chem. C 118
doped thin film, and 5% Zn-doped film slightly deviates from 5908–17

6
J. Phys.: Condens. Matter 33 (2021) 015602 K Soni et al

[2] Onozuka T et al 2017 Reversible changes in resistance of per- [22] Garcı́a-Muñoz J L, Suaaidi M, Martı́nez-Lope M J and Alonso
ovskite nickelate NdNiO3 thin films induced by fluorine J A 1995 Influence of carrier injection on the metal-insulator
substitution ACS Appl. Mater. Interfaces 9 10882–7 transition in electron- and hole-doped R1−x Ax NiO3 per-
[3] Scherwitzl R, Zubko P, Lezama I G, Ono S, Morpurgo A F, ovskites Phys. Rev. B 52 13563–9
Catalan G and Triscone J-M 2010 Electric-field control of [23] Chandra M, Rana R, Aziz F, Khare A, Rana D S and
the metal-insulator transition in ultrathin NdNiO3 films Adv. Mavani K R 2013 Competing effects of Mn-doping
Mater. 22 5517–5520 and strain on electrical transport of NdNi1−x Mnx O3
[4] Piamonteze C, Tolentino H C N, Ramos A Y, Massa N E, (0x0.10) thin films J. Phys. D: Appl. Phys. 46
Alonso J A, Martı́nez-Lope M J and Casais M T 2005 Short- 415305
range charge order in RNiO3 perovskites (R = Pr, Nd, Eu, [24] Chandra M, Das S, Aziz F, Prajapat M and Mavani K R 2017
Y) probed by x-ray-absorption spectroscopy Phys. Rev. B 71 Induced metal-insulator transition and temperature indepen-
012104 dent charge transport in NdNiO3−δ thin films J. Alloys
[5] Wang L et al 2016 Oxygen vacancy induced room-temperature Compd. 696 423–7
metal–insulator transition in nickelate films and its potential [25] Chang L, Wang L, You L, Zhou Y, Fang L, Wang S
application in photovoltaics ACS Appl. Mater. Interfaces 8 and Wang J 2016 Band gap tuning of nickelates for
9769–76 photovoltaic applications J. Phys. D: Appl. Phys. 49
[6] Zhang Z et al 2018 Perovskite nickelates as electric-field sensors 44LT02
in salt water Nature 553 68–72 [26] Lian X K, Chen F, Tan X L, Chen P F, Wang L F,
[7] Giovannetti G, Kumar S, Khomskii D, Picozzi S and van den Gao G Y, Jin S W and Wu W B 2013 Anisotropic-strain-
Brink J 2009 Multiferroicity in rare-earth nickelates RNiO3 controlled metal-insulator transition in epitaxial NdNiO3
Phys. Rev. Lett. 103 156401 films grown on orthorhombic NdGaO3 substrates Appl. Phys.
[8] Chaloupka J and Khaliullin G 2008 Orbital order and possible Lett. 103 172110
superconductivity in LaNiO3 /LaMO3 superlattices Phys. Rev. [27] Dong Y, Xu H, Luo Z, Zhou H, Fong D D, Wu W and Gao C
Lett. 100 016404 2017 Effect of gate voltage polarity on the ionic liquid gating
[9] Rüegg A and Fiete G A 2011 Topological insulators from com- behavior of NdNiO3 /NdGaO3 heterostructures APL Mater. 5
plex orbital order in transition-metal oxides heterostructures 051101
Phys. Rev. B 84 201103 [28] Heo S, Oh C, Eom M J, Kim J S, Ryu J, Son J and Jang H M 2016
[10] Zhou J-S, Goodenough J B and Dabrowski B 2003 Transition Modulation of metal-insulator transitions by field-controlled
from curie-weiss to enhanced pauli paramagnetism in RNiO3 strain in NdNiO3 /SrTiO3 /PMN-PT (001) heterostructures
(R= La, Pr, Gd) Phys. Rev. B 67 020404 Sci. Rep. 6 22228
[11] Catalano S, Gibert M, Fowlie J, Íñiguez J, Triscone J-M and
[29] Liu J et al 2013 Heterointerface engineered electronic and
Kreisel J 2018 Rare-earth nickelates RNiO3 : thin films and
magnetic phases of NdNiO3 thin films Nat. Commun. 4 2714
heterostructures Rep. Prog. Phys. 81 046501
[30] Breckenfeld E, Chen Z, Damodaran A R and Martin L W
[12] Medarde M L 1997 Structural, magnetic and electronic proper-
2014 Effects of nonequilibrium growth, nonstoichiome-
ties of RNiO3 perovskites (R = rare earth) J. Phys.: Condens.
try, and film orientation on the metal-to-insulator transi-
Matter 9 1679
tion in NdNiO3 thin films ACS Appl. Mater. Interfaces 6
[13] Canfield P C, Thompson J D, Cheong S-W and Rupp L W 1993
22436–44
Extraordinary pressure dependence of the metal-to-insulator
transition in the charge-transfer compounds NdNiO3 and [31] Alonso J A, Martı´nez-Lope M J and Hidalgo M A 1995 Hole
PrNiO3 Phys. Rev. B 47 12357–60 and electron doping of RNiO3 (R = La, Nd) J. Solid State
[14] Torrance J B, Lacorre P, Nazzal A I, Ansaldo E J and Nieder- Chem. 116 146–56
mayer C 1992 Systematic study of insulator-metal transitions [32] Shi J, Zhou Y and Ramanathan S 2014 Colossal resistance
in perovskites RNiO3 (R=Pr, Nd, Sm, Eu) due to closing of switching and band gap modulation in a perovskite nickelate
charge-transfer gap Phys. Rev. B 45 8209–12 by electron doping Nat. Commun. 5 4860
[15] Catalan G 2008 Progress in perovskite nickelate research Phase [33] Álvarez I, Veiga M L and Pico C 1998 Metal–insulator
Transit. 81 729–49 transitions induced by doping in LaNiO3 : LaNi0.95 M0.05 O3
[16] Torrance J B, Lacorre P, Asavaroengchai C and Metzger R (M=Mo, W, Sb, Ti, Cu, Zn) perovskites J. Solid State Chem.
M 1991 Why are some oxides metallic, while most are 136 313–9
insulating? Phys. C 182 351–64 [34] Blasco J and Garcı́a J 1994 Structural, magnetic and electri-
[17] Medarde M, Dallera C, Grioni M, Delley B, Vernay F, Mesot J, cal properties of NdNi1−x Fex O3 and NdNi1−x Cox O3 systems
Sikora M, Alonso J A and Martı́nez-Lope M J 2009 Charge J. Phys. Chem. Solids 55 843–52
disproportionation in RNiO3 perovskites (R= rare earth) [35] Pérez J, Stankiewicz J, Blasco J, Castro M and Garcı́a J 1996
from high-resolution x-ray absorption spectroscopy Phys. The metal–insulator transition in NdNi1−x Cux O3 perovskites
Rev. B 80 245105 J. Phys.: Condens. Matter 8 10393
[18] Mikheev E, Hauser A J, Himmetoglu B, Moreno N E, Janotti [36] Chandra M, Das S, Aziz F, Tripathi S and Mavani K R 2015
A, Wallede C G V and Stemmer S 2015 Tuning bad metal Large effects of epitaxial tensile strain on electrical transport
and non-Fermi liquid behavior in a Mott material: rare-earth of Mn-doped NdNiO3 thin films Solid State Commun. 219
nickelate thin films Sci. Adv. 1 e1500797 16–20
[19] Chandra M, Aziz F, Tripathi S, Rana R, Rana D S and Mavani [37] Sun Y H et al 2008 Electric and magnetic modulation of fully
K R 2014 Contrasting effects of doping on insulating and strained dead layers in La0.67 Sr0.33 MnO3 films Phys. Rev.
metallic states of NdNi1−x Mnx O3 thin films J. Appl. Phys. B 78 024412
115 093701 [38] Massa N E, Ramos A Y, Tolentino H C N, Sousa-Neto N M,
[20] Yao D, Shi L, Zhou S, Liu H, Zhao J, Li Y and Wang Y 2016 Fonseca J Jr and Alonso J A 2015 Temperature and high-
Tuning the metal-insulator transition via epitaxial strain and pressure dependent x-ray absorption of SmNiO3 at the Ni K
Co doping in NdNiO3 thin films grown by polymer-assisted and Sm L3 edges Mater. Res. Express 2 126301
deposition J. Appl. Phys. 119 035303 [39] Acosta-Alejandro M, de León J M, Medarde M, Lacorre P,
[21] Wang L, Chang L, Yin X, Rusydi A, You L, Zhou Y, Fang L Konder K and Montano P A 2008 Local lattice struc-
and Wang J 2017 Localization-driven metal–insulator transi- ture change in PrNiO3 across the metal-insulator transition:
tion in epitaxial hole-doped Nd1−x Srx NiO3 ultrathin films J. x-ray absorption near-edge structure spectroscopy and ab
Phys.: Condens. Matter 29 025002 initio calculations Phys. Rev. B 77 085107

7
J. Phys.: Condens. Matter 33 (2021) 015602 K Soni et al

[40] Kumar Y, Choudhary R J, Sharma S K, Knobel M and Kumar [44] Hauser A J, Mikheev E, Moreno N E, Cain T A, Hwang J, Zhang
R 2012 Strain dependent stabilization of metallic paramag- J Y and Stemmer S 2013 Temperature-dependence of the Hall
netic state in epitaxial NdNiO3 thin films Appl. Phys. Lett. coefficient of NdNiO3 thin films Appl. Phys. Lett. 103 182105
101 132101 [45] Ha S D, Jaramillo R, Silevitch D M, Schoofs F, Kerman K,
[41] Girardot C, Kreisel J, Pignard S, Caillault N and Weiss F Baniecki J D and Ramanathan S 2013 Hall effect measure-
2008 Raman scattering investigation across the magnetic ments on epitaxial SmNiO3 thin films and implications for
and metal-insulator transition in rare earth nickelate RNiO3 antiferromagnetism Phys. Rev. B 87 125150
(R= Sm, Nd) thin films Phys. Rev. B 78 104101 [46] Phanindra V E, Agarwal P and Rana D S 2018 Terahertz spectro-
[42] Shao T, Qi Z, Wang Y, Li Y, Yang M and Hu C 2017 Metal- scopic evidence of non-Fermi-liquid-like behavior in struc-
insulator transition in epitaxial NdNiO3 thin film: a structural, turally modulated PrNiO3 thin films Phys. Rev. Mater. 2
electrical and optical study Appl. Surf. Sci. 399 346–50 015001
[43] Jaramillo R, Ha S D, Silevitch D M and Ramanathan S [47] Phanindra V E, Das S, Kumar K S, Agarwal P, Rana R and Rana
2014 Origins of bad-metal conductivity and the insula- D S 2017 Tuning the terahertz low-energy charge dynamics
tor–metal transition in the rare-earth nickelates Nat. Phys. 10 by simultaneous effect of epitaxial and anisotropic strain in
304–7 PrNiO3 thin films Phys. Rev. B 95 085114