Induced metal-insulator transition and temperature independent charge transport in NdNiO3-δ thin films

Induced metal-insulator transition and temperature independent charge transport in NdNiO3-δ thin films

Accepted Manuscript Induced metal-insulator transition and temperature independent charge transport in NdNiO3-δ thin films Mahesh Chandra, Sarmistha D...

1MB Sizes 11 Downloads 92 Views

Accepted Manuscript Induced metal-insulator transition and temperature independent charge transport in NdNiO3-δ thin films Mahesh Chandra, Sarmistha Das, Fozia Aziz, Manoj Prajapat, K.R. Mavani PII:

S0925-8388(16)33590-3

DOI:

10.1016/j.jallcom.2016.11.122

Reference:

JALCOM 39618

To appear in:

Journal of Alloys and Compounds

Received Date: 22 September 2016 Revised Date:

7 November 2016

Accepted Date: 8 November 2016

Please cite this article as: M. Chandra, S. Das, F. Aziz, M. Prajapat, K.R. Mavani, Induced metalinsulator transition and temperature independent charge transport in NdNiO3-δ thin films, Journal of Alloys and Compounds (2016), doi: 10.1016/j.jallcom.2016.11.122. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Induced metal-insulator transition and temperature independent charge transport in NdNiO3-δ thin films Mahesh Chandra1, Sarmistha Das2, Fozia Aziz1, Manoj Prajapat2, K.R. Mavani1,3,* 1

2

RI PT

Discipline of Physics, Indian Institute of Technology (IIT) Indore, Simrol, Khandwa Road, M. P. 453 552, India Indian Institute of Science Education and Research (IISER) Bhopal, M.P. 462023, India

3

Metallurgical Engineering and Materials Science, Indian Institute of Technology (IIT) Indore, Simrol, Khandwa Road, M. P. 453 552, India

SC

Abstract

The ground state of RNiO3 (R = Rare earth ion) films can be influenced by thickness,

M AN U

strain and oxygen content. We have deposited two series of epitaxial thin films of NdNiO3-δ (NNO): one with variation in thickness (5 nm – 16 nm) and another with variation in oxygen content and fixed thickness. In spite of these variations, the films show epitaxial growth on (LaAlO3)0.3(Sr2AlTaO6)0.7– [(LSAT) (001)] single-crystal substrates. Electrons and holes, both carry charge for transport in NNO films. The Hall resistance measurements show switching of majority charge carriers from holes to electrons as temperature is reduced.

TE D

Temperature independent resistivity (in µΩ.cm) is observed over a wide temperature range around 300 K. These results reveal that the ground state of NdNiO3-δ can be modified in order to achieve the required temperature coefficient of resistance at room temperature and a fine

EP

control can be achieved in combination with optimal oxygen content at thickness close to

AC C

dimensionality crossover.

*Corresponding author: Dr. Krushna R. Mavani, Associate Professor Indian Institute of Technology Indore Ph:+91-731-2438923 E-mail: [email protected] 1

ACCEPTED MANUSCRIPT Introduction Dimensionality crossover from three (3D) to two (2D) and defect induced disorder give rise to profound effects in materials [1-3]. A dimensionality crossover has been found to induce orbital reconstruction or orbital polarization leading to diminishing carrier hopping in

RI PT

z-direction [4,5], transition from metal to Mott insulator [6] and strong localization in combination with strain in ultrathin films [7]. Moreover, Anderson localization becomes favorable in a 2D disordered system, where extended states are suppressed and get localized due to strong disorder as compared to a 3D system [8,9]. In general in oxides, disorder is

SC

inevitable due to the presence of oxygen vacancies. This localization may not alone be Anderson type, but Anderson-Mott localization as a combined effect [9]. Thus, a disorder and dimensionality (induced by oxygen variation and decreasing film thickness) may give rise to

M AN U

altered electronic ground states in thin film materials.

Among strongly correlated oxides, RNiO3 (R=Rare earth ion) compounds are famous for a temperature driven first-order metal to insulator (MI) transition. In the insulating state, they are characterized as charge transfer type insulators with a small gap between O 2p and Ni 3d bands. Due to this small charge transfer gap, their electronic state is very sensitive to

TE D

external perturbations like pressure, strain, doping, vacancy and dimensionality. The charge transfer gap itself mostly varies as a function of these parameters in RNiO3 system [10-16]. A metallic state is characterized by a positive temperature coefficient of resistance (α) and the sign of α may be altered by external parameters. Thus, at the verge of the external

EP

perturbation induced transitions in ground state, it may be possible to switch the electronic state from metallic to insulating and vice-versa. RNiO3 system has both the types of charge

AC C

carriers. The Fermi surface, containing both electron and holes, is observed in some alloys, semimetals and compounds, where a state of temperature independent charge transport has been achieved [17-20]. The materials with a temperature independent transport have many applications, such as in high precision electronic circuits (as a resistor), thermoelectric devices and sensing application in automobile [21-23]. With the motivation of exploring the influence of dimensionality and disorder on the electronic state of RNiO3 thin films, we have investigated two series of epitaxial NdNiO3-δ (NNO) thin films deposited on (LaAlO3)0.3(Sr2AlTaO6)0.7– [(LSAT) (001)]. In this report, we show that the temperature independent conductive state of epitaxial NNO films and tuning of electronic properties can be achieved at certain thickness-range and oxygen content.

2

ACCEPTED MANUSCRIPT Experimental NNO thin films were deposited using pulsed laser deposition, where KrF eximer laser (λ=248 nm) was used for ablating the target material. For all depositions, the laser energy was set at 310 mJ and pulse repetition rate was 4 Hz. During deposition, oxygen partial pressure was kept at 40 Pa. Target to substrate distance was 4.5 cm and substrate temperature was

RI PT

maintained at 720°C during deposition. Keeping the above parameters fixed in all depositions, in series-A (with fixed thickness of 12 nm) the thin films were post annealed for 0 min (unannealed), 3 min and 5 min in oxygen partial pressure of 1000 Pa, whereas in series-B the thickness was varied (from ~ 5 nm to 16 nm) with a constant oxygen annealing

SC

time of 3 min at 1000 Pa. For the present study, we have chosen the thickness range across the dimensionality crossover. The growth rate of few of the films was determined by X-ray

estimated for the rest of the films. Results and discussions

M AN U

reflectivity (XRR) measurements, and then based on this growth rate the thickness was

The pseudo-cubic lattice parameter of NdNiO3 is 3.80Å, whereas that of LSAT is 3.86 Å, hence these thin film experiences an in-plane tensile strain on LSAT (001) substrate. The X-Ray Diffraction (XRD) patterns and Reciprocal space maps (RSMs) for both the series

TE D

are shown in Figs. 1 and 2, respectively. It is noteworthy that, in spite of oxygen annealing for different duration and thickness variation in both the series, all the films under study are phase pure, epitaxial, and coherently strained. Thus, the in-plane parameters and hence the state of strain is almost same for all these films, which provides an opportunity to see the

EP

effects of thickness and oxygen variations in NNO system without any ambiguity. For series-A, the XRD peaks shifted towards higher 2θ with increased oxygen annealing time, indicating a decrease in the unit cell volume (Fig. 1(a)). Here an oxygen vacancy is

AC C

accompanied by a reduction in the valence state of nearby metal ion, i.e., from Ni3+ ions (ionic radii = 56 pm) to more stable state Ni2+ ions (ionic radii = 69 pm). Therefore, the unannealed film which contains larger amount of vacancy has largest unit cell volume and 5 min annealed film with more oxygen has smallest unit cell volume. The XRD patterns of series-B (Fig. 1(b)) show a very systematic shift along with a systematic broadening in the peaks with increasing thickness from 5 nm to 16 nm. Firstly, the peak shifting towards higher 2θ indicates a reduction in the unit cell volume. Hence, the surface to volume ratio increases with the reduced thickness. As the oxygen vacancy formation energy is smaller at the surface and interface [24], the thinnest film should have more density of oxygen vacancies. These vacancies increase the size of Ni ions as explained 3

ACCEPTED MANUSCRIPT earlier. The above argument is also supported in the view of broadening of peaks and change in morphology with thickness. Peak broadening mainly occurs due to the variance in the grain size. Sharper peak indicates higher crystallinity (and larger grain size), whereas broader peaks indicated lower crystallinity (and smaller grain size) in the films. The thick films are more crystalline compare to thin films. Additionally, Scanning electron Microscopy (SEM)

RI PT

images (Fig. 2) clearly show an island-like growth for 5 nm film, whereas 16 nm film has a smooth and uniformly deposited surface. Owing to the larger surface area and the island formation, the reactivity of the surface on thinnest film will be large resulting into creation of oxygen vacancies.

SC

Resistivity versus temperature plots for both series of films are shown in Fig. 3 (a,b). Very systematic variations are observed in resistivity of the films while varying oxygen content. The resistivity increases with increasing oxygen vacancy in metallic state whereas it

M AN U

decreases in insulating state with flattening of the curve. The observed trends of resistivity agree with the earlier reports to show that the oxygen vacancies act as scattering centers for charge carriers in the metallic state, but they enhance density of states in the insulating state by acting as a donor impurity [25, 26]. Moreover, the conversion from Ni3+ to Ni2+ decreases the band-width and opens up the charge transfer gap, [11, 27]. In present case, the ratio of 2+ and 3+ valance state of Ni is altered by the presence of oxygen vacancies. The increased

TE D

conversion to Ni2+ at higher oxygen vacancies diminishes the metallic state. Thus, the films with 0 min oxygen annealing in series-A does not show any metallic nature till room temperature, rather a constant resistivity appears at higher temperatures.

EP

Unlike series-A, the resistivity of series-B shows a decrease with increase in thickness from 5 nm to 12 nm irrespective of metallic/insulating state. Let us note that the chosen

AC C

thickness range of 5 nm to 16 nm lies across the dimensionality crossover from 2D to 3D charge transport [28]. Due to higher thickness, the crystallinity is better (Fig. 1) and charge transport is three dimensional in 16 nm film. Thus, at the thickness of 16 nm, a less broad MI transition is induced and a crossover is observed with other resistivity curves in the figure. One notable feature is a temperature independent state of resistivity in a very wide

temperature range (inset Figs. 3 (a, b)) for two of these films. The film with thickness of 9 nm (of series B) and unannealed 12 nm thin film (of series-A) show a constant resistivity. The normalized resistance varies by ±0.003 for 9nm film and ±0.02 for unannealed film (shaded region inset figures). Also it is notable here that the temperature coefficient of resistance tunes quite systematically with variation of thickness or oxygen content (Fig. 3(c)). Such temperature independent behavior of resistance has been observed earlier in alloys consisting 4

ACCEPTED MANUSCRIPT of Cu, Ni and Mn [17], and in Cu3NMx (M-Cu, Ag, Au) compounds and conducting polymers [18-20, 29]. This behavior has never been reported near room temperature for other perovskites and hence unique to be observed in RNiO3-type thin films. Also, it should be noted that, this state has been observed for a thickness range at dimensionality crossover. Earlier reports of oxygen variations do not show any such temperature independent resistive

RI PT

state in these films for 3D electronic transport (at higher thickness), and the films with 2D electronic transport anyway show insulating state most of the times [for examples see refs. 9,11,25,26].

There are two parameters which may bring such temperature independent conductive

SC

state: a) Anderson localization, and b) a gradually opening of band gap in 2D structure. For clearer picture, we tried fitting variable-range-hopping model (ρ=ρ0exp(T0/T)1/d+1, where d is dimensionality of the system), as well as activation behavior (ρ=ρ0exp(Ea/kT), where Ea is

M AN U

activation energy), to the resistivity of insulating film in series-B. Only activation behavior could fit well to the resistivity curves indicating that the insulating behavior is not influenced by a disorder in the 2D film (5 nm). Thus, as shown in Fig. 3(d), the thickness variation resulting into dimensionality crossover gives rise to creation of band gap. Such opening of band-gap has been earlier observed by in-situ photoemission experiments on PrNiO3 thin

TE D

films [12]. Considering that NNO is semi-metallic, the temperature independence of ρ can be understood in terms of a fine balance between carrier mobility and carrier concentration, which are mainly responsible for temperature dependence of resistivity. These two quantities have an opposite trend with temperature in a semimetallic state, where the band gap is zero

EP

(as shown in Fig. 3(d)) and a material has properties of both, metal and semiconductor. So, with increasing temperature the mobility decreases (like in metals) and carrier concentration

AC C

increases (like in semiconductors).

One can understand the role of oxygen vacancies by comparing the temperature

independent transport of unannealed 12 nm film in series-A with that of the 9 nm thin film of series-B. Oxygen vacancies are known to be a source of disorder in oxides which can transform the ground state as explained earlier. Although 12 nm film may have 3D transport, it is at the verge of dimensionality cross over and therefore the metallic state is affected by the induced disorder which drives the system towards insulating state in series-A. As variable-range-hopping model does not fit to the low-temperature insulating state in present case of 12 nm film, the high-temperature semimetallic sate with constant resistivity can be

5

ACCEPTED MANUSCRIPT said to be achieved by a decrease in the oxygen content at the thickness of 12 nm, at the verge of dimensionality crossover. In order to get further insight into the carriers involved in the charge conduction, Hall Effect measurements were performed for 5 nm (series-B), 9 nm (series-B) and 12 nm (seriesA) films (Fig. 4). These chosen films have different features like 2D transport, temperature

RI PT

independent resistivity and a room-temperature metallic state, respectively. The positive slope of the Hall resistance versus field curve at higher temperature shows that the majority charge carriers are holes at room temperature. A switching of majority charge carriers to electrons is observed below 200K. These results are in agreement with the earlier reports on

SC

RNiO3 [26,30].The switching in the sign of Hall coefficient is also observed in other semimetallic systems [31]. NdNiO3 is a multiple band system, where both electrons and holes participate in conduction and for such systems the Hall coefficient is given by:

M AN U

మ ା௣ఓ మ ି௡ఓ೙ ೛

ܴு = ௘(௡ఓ

మ ೙ ା௣ఓ೛ )

(1)

Where n and p is the electron and hole density respectively, and µ is their respective mobility. Now, looking upon the values of RH (Table I) for three films with different thickness, an anomaly is consistently observed for all the films. The value of RH first increases going

TE D

from 300 K to 250 K and then decreases for 200 K. The increase in RH indicated a decrease in majority charge carriers when only one type of carrier are involved but in preset case NdNiO3-δ is a two band system therefore the change in the value to RH is a collective response of the both type of carrier concentration and their respective mobility. This anomaly

EP

indicates a competition between these various parameters of Eqn. 1 in this temperature range. This inconsistent change in RH is not profound except in 9 nm thin film, which is situated at

AC C

the dimensionality crossover and shows a constant resistance in the same temperature range. Therefore, the competition between the two types of carrier densities as well as their mobility play roles in keeping resistivity constant for the thicknesses near dimensionality crossover. Conclusions

Two series of epitaxial NdNiO3-δ films have been studied by varying thickness and

oxygen content. Systematic variation in out-of-plane parameter is observed due to unit cell volume expansion as a result of oxygen annealing. The majority charge carriers show a switching from holes to electrons as temperature decreases below 200 K. The resistivity of series-A shows systematic trends with varying oxygen content, clearly defining the role of oxygen vacancies in metallic and insulating temperature regions. With thickness variation, 6

ACCEPTED MANUSCRIPT the series-B of films shows an electronic transition from metallic to complete insulating state in 5 nm film due to dimensionality crossover and opening of charge transfer gap. By variations in thickness and oxygen content, a semi-metallic state has been achieved for two films where the resistivity shows constant values over a wide range of temperatures around room temperature. This kind of behavior is rarely observed for perovskites. Such materials

RI PT

may find applications as precision resistor in electronic devices, circuits and sensors where a temperature independent resistance is desired. References

1. Defects and Surface-Induced Effects in Advanced Perovskites edited by Gunnar

SC

Borstel, Andris Krumins, Donats Millers. Springer Netherlands, ISBN 978-94-0114030-0

Rev. Lett. 106, 246403 (2011).

M AN U

2. R. Scherwitzl, S. Gariglio, M. Gabay, P. Zubko, M. Gibert, and J.-M. Triscone, Phys.

3. Man Gu, Stuart A. Wolf, Jiwei Lu, Adv. Mat. 1,7 (2014).

4. Jian Liu, M. Kareev, D. Meyers, B. Gray, P. Ryan, J.W. Freeland, and J. Chakhalian, Phys. Rev. Lett. 109, 107402, (2012).

5. H. K. Yoo, S. I. Hyun, L. Moreschini, Y. J. Chang, D. W. Jeong, C. H. Sohn, Y. S.

TE D

Kim, H.-D. Kim, A. Bostwick, E. Rotenberg, J. H. Shim, T. W. Noh. (2013) arXiv:1309.0710 [cond-mat.str-el].

6. Raoul Scherwitzl , Pavlo Zubko , I. Gutierrez Lezama , Shimpei Ono, Alberto F.

(2010).

EP

Morpurgo , Gustau Catalan , and Jean-Marc Triscone, Adv. Mater., 22, 5517–5520,

7. Junwoo Son, Pouya Moetakef, James M. LeBeau, Daniel Ouellette, Leon Balents, S.

AC C

James Allen, and Susanne Stemmer, Appl. Phys. Lett. 96, 062114, (2010). 8. P. A. Lee, and T. V. Ramakrishnan, Rev. Mod. Phys. 57, 287, 337 (1985). 9. V.F. Gantmakher, Electrons and disorder in solids, Clarendon press, Oxford 2005, ISBN 0–19–856756–1 (Hbk.) 978–0–19–856756–1

10. Torrance J., Lacorre P., Nazzal A., Ansaldo E., and Niedermayer C., Phys. Rev. B 45, 8209–12, (1992). 11. Le Wang, Sibashisa Dash, Lei Chang, Lu You, Yaqing Feng, Xu He, Kui-juan Jin, Yang Zhou,Hock Guan Ong, Peng Ren, Shiwei Wang, Lang Chen, and Junling Wang. Appl. Mater. Interfaces, 8, 9769−9776 (2016).

7

ACCEPTED MANUSCRIPT 12. Enju

Sakai, Kohei

Yoshimatsu, Masatomo

Tamamitsu, Koji

Horiba,Atsushi

Fujimori, Masaharu Oshima, Hiroshi Kumigashira, Appl. Phys. Lett. 104, 171609, (2014). 13. Ashutosh Tiwari, C. Jin, J. Narayan, Appl. Phys. Lett. 80, 4039, (2002).

14. Catalan G., Bowman R. M. and Gregg J. M., Phys. Rev. B 62, 7892–900(2000). J. Appl. Phys. 25, 25-31 (2004)

RI PT

15. P. Laffez, O. I. Lebedev, P. Ruello, R. Desfeux, G. Banerjee and F. Capon, Eur. Phys.

16. K. Okazaki, T. Mizokawa, A. Fujimori, E. V. Sampathkumaran, M. J. Martinez-Lope, and J. A. Alonso, Phys. Rev. B, 67, 073101 (2003).

0871707268).

SC

17. Davis J. R., (2001) Copper and Copper Alloys, ASM International (ISBN:

18. Ji A., Li C., Cao Z., Appl. Phys. Lett. 89, 252120 (2006).

M AN U

19. Lu N., Ji A., Cao Z., Sci. Rep. 3, 3090 (2013).

20. Lei Ding, Cong Wang, Lihua Chu, Jun Yan,Yuanyuan Na, Qingzhen Huang Xiaolong Chen. Appl. Phys. Lett. 99, 251905 (2011). 21. B. Fu and L. Gao, Scr. Mater. 55, 521 (2006).

22. J. Horo, P. G. Harne, B. B. Nayak, and S. Vitta, Mater. Sci. Eng., B 107, 53 (2004).

TE D

23. Y.T. Kim, Appl. Phys. Lett., 70, 20 (1997).

24. Boyd W. Veal, Seong Keun Kim, Peter Zapol, Hakim Iddir, Peter M. Baldo & Jeffrey A. Eastman, Nat. Commun., 7, 11892 (2016). 25. Sieu D. Ha, Ulrich Vetter, Jian Shi, Shriram Ramanathan, Appl. Phys. Lett. 102,

EP

183102 (2013).

26. Sieu D. Ha, R. Jaramillo, D. M. Silevitch, Frank Schoofs, Kian Kerman, John D.

AC C

Baniecki, Shriram Ramanathan, Phys. Rev. B. 87, 125150 (2013). 27. Álvarez I, Veiga M. and Pico C., J. Solid State Chem. 136, 313–9 (1998)

28. Le Wang, Sheng Ju, Lu You, Yajun Qi, Yu-wei Guo, Peng Ren, Yang Zhou & Junling Wang, Sci. Rep. 5, 18707 (2015).

29. Kunmo Chu, Sung-Chul Lee, Sangeui Lee, Dongearn Kim, Changyoul Moon, SungHoon Park, Nanoscale, 7, 471 (2015). 30. Hauser A. J., Mikheev E., Moreno N. E., Cain T. A., Hwang J., Zhang J. Y. and Stemmer S., Appl. Phys. Lett. 103, 182105 (2013).

31. Imai M., Abe H. and Yamada K., Inorg. Chem. 43, 5186–8, (2004).

8

ACCEPTED MANUSCRIPT Table I: Temperature dependent Hall coefficient (RH) for three chosen films of different thickness. 300 K

250 K

200 K

150 K

100 K

5 nm (B) 9 nm (B) 12 nm (A)

2.16×10-4 5.87×10-3 8.8×10-5

3.2×10-4 9.5×10-3 5.79×10-4

-6.27×10-4 1.15×10-3 7.7×10-5

-2.43×10-4 ---3.084×10-4

---4.2×10-3 -1.07×10-3

RI PT

RH (cm3/C)

Figure captions:

SC

Figure 1: XRD peaks (002) for NNO thin films (a) with different oxygen annealing and (b) with different thickness.

M AN U

Figure 2: RSMs around 301 peak for (a) Unannealed film (series-A) (b) 5 min annealed film (series-A) and (c) 5 nm (series-B) and (d)16 nm film (series-B). SEM images for (e) 5 nm film (series-B) and (f) 16 nm film (series-B) Figure 3: Resistivity versus temperature plots for NNO thin films with, (a) for series-A and (b) for series-B. The insets are the magnified curve for films showing constant resistance in respective series. (c) The temperature coefficient of resistivity (α) of films in series A and B. (d) Schematic shows an opening of charge transfer gap (∆) by decreasing thickness.

TE D

Figure 4: Hall resistance versus magnetic field plot for chosen NNO/LSAT films: 5 nm (series-B), 9 nm (series-B) and 12 nm (series-A).

AC C

EP

*******

9

ACCEPTED MANUSCRIPT Table I: Temperature dependent Hall coefficient (RH) for three chosen films of different thickness. 300 K

250 K

200 K

150 K

100 K

5 nm (B) 9 nm (B) 12 nm (A)

2.16×10-4 5.87×10-3 8.8×10-5

3.2×10-4 9.5×10-3 5.79×10-4

-6.27×10-4 1.15×10-3 7.7×10-5

-2.43×10-4 ---3.084×10-4

---4.2×10-3 -1.07×10-3

AC C

EP

TE D

M AN U

SC

RI PT

RH (cm3/C)

ACCEPTED MANUSCRIPT

46

2

48

RI PT

48

AC C

EP

TE D

M AN U

SC

2

Intensity (a.u.)

NNO(002)

16 nm 12 nm 9 nm 5 nm

0 min annealing 3 min annealing 5 min annealing

46

NNO(002)

(b)

LSAT(002)

Intensity (a.u.)

(a)

LSAT(002)

Figure 1

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 2

ACCEPTED MANUSCRIPT





0

300

(c)

1

T(K)

300

5 nm 9 nm 12 nm 16 nm

T(K)

200

300

TE D EP

AC C

200

1.10

250

300

200

T(K)

300

unannealed

NdNiO3/LSAT(001)

1.05

1.00

M AN U

100

2.5

150

100

300

T(K)

3.0

1.0

0 0

200

3.5

1.5

4

NN-LS(OD)-H

0.003

-3

2.0

16

(230K-350 K)

10

0

12

(b)

1.02

4.0

8

Thickness (nm)

-5

9 nm

8

RI PT

200



5

5 min annealing 3 min annealing Unannealed

200

-4

6

SC

10

100

4

0.02

100



-2

50

2

(140K-302 K)

4 -4

10

Annealing time (min)

T(K)

10

0

0



(cm)

-3

10

(cm)

(a)

1.15 1.10 1.05 1

  

-2

10



Figure 3

(d)

ACCEPTED MANUSCRIPT Figure 4

0.5

RI PT

5 nm 0

9 nm 2 0 -2

M AN U

4

TE D

r(B)-r(-B)/2 ()

-1.0

SC

300 K 250 K 200 K 150 K 100 K

-0.5

12 nm ( 5 min annealed)

0.3

EP

0 -0.3

AC C

-0.6

0

1

2

B(T)

3

4

5

ACCEPTED MANUSCRIPT

Highlights:

AC C

EP

TE D

M AN U

SC

RI PT

1. Resistivity varies very systematically with increasing oxygen content in NdNiO3-δ films 2. Dimensionality crossover induced metal-insulator transition is observed by reducing the thickness of the films to 5 nm. 3. A temperature independent charge transport is observed in a very wide temperature range in some of the films 4. Temperature coefficient of resistance varies systematically, and can be controlled by thickness/oxygen variation.