Crossover between weak anti localization and weak localization by co doping and annealing in gapless pbpdo2 and spin gapless co doped pbpdo2

Crossover between weak anti-localization and weak localization by Co doping andannealing in gapless PbPdO2 and spin gapless Co-doped PbPdO2 S... In magnetoconductance curves, we observed

Crossover between weak anti-localization and weak localization by Co doping and

annealing in gapless PbPdO2 and spin gapless Co-doped PbPdO2

S M Choo, K J Lee, S M Park, J B Yoon, G S Park, C.-Y You, and M H Jung

Citation: Applied Physics Letters 106, 172404 (2015); doi: 10.1063/1.4919452

View online: http://dx.doi.org/10.1063/1.4919452

View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/106/17?ver=pdfcov

Published by the AIP Publishing

Articles you may be interested in

Magnetic versus nonmagnetic ion substitution effects in gapless semiconductor PbPdO2

Appl Phys Lett 106, 072406 (2015); 10.1063/1.4913301

Electronic structure of the spin gapless material Co-doped PbPdO2

J Appl Phys 114, 103709 (2013); 10.1063/1.4821039

Annealing effect on surface morphology and electrical transport of PbPdO2 and Pb(Pd,Co)O2

J Appl Phys 113, 014904 (2013); 10.1063/1.4772723

Weak localization and percolation effects in annealed In2O3-ZnO thin films

AIP Advances 1, 032149 (2011); 10.1063/1.3635375

Weak anti-localization of the two-dimensional electron gas in modulation-doped Al x Ga 1 − x N ∕ GaN

heterostructures with two subbands occupation

Appl Phys Lett 85, 3125 (2004); 10.1063/1.1803949

Crossover between weak anti-localization and weak localization by Co

S M.Choo,1K J.Lee,1S M.Park,1J B.Yoon,2G S.Park,1C.-Y.You,3and M H.Jung1,a)

1

Department of Physics, Sogang University, Seoul 121-742, South Korea

2

Department of Electrical and Computer Engineering and NUSNNI, National University of Singapore,

Singapore 117576, Singapore

3

Department of Physics, Inha University, Incheon 402-751, South Korea

(Received 20 January 2015; accepted 20 April 2015; published online 30 April 2015)

The magnetotransport properties of Pb(Pd,Co)O2 and PbPdO2 thin films were investigated In

magnetoconductance curves, we observed a crossover between weak anti-localization (WAL) and

weak localization (WL) depending on the annealing and Co doping in PbPdO2thin films For the

Pb(Pd,Co)O2case showing WAL signals, theex-situ annealing weakens the Pd-O hybridization by

stabilizing Co3þstates and generating Pd1þstates, instead of Pd2þ, so that the spin-orbit coupling

(SOC) strength is significantly reduced It causes the dominant magnetotransport mechanism

change from WAL to WL This annealing effect is compared with the PbPdO2case, which

pos-sesses WL signals The annealing process stabilizes the oxygen states and enhances the Pd-O

hybridization, and consequently the SOC strength is enhanced Our experimental results are well

explained by the Hikami-Larkin-Nagaoka theory in terms of two important physical parameters;

SOC strength-related a and inelastic scattering lengthl/.V C 2015 AIP Publishing LLC

[http://dx.doi.org/10.1063/1.4919452]

Gapless semiconductors,1,2that have zero gap between

the conduction and valence bands at the Fermi level, are

exciting materials to show unique electronic properties

caused by their exotic band structure They are extremely

sen-sitive to external influences such as chemical doping, lattice

imperfection, temperature, and magnetic field.1,35Recently,

a theoretical prediction of a spin gapless semiconductor has

been developed by inducing spin degree of freedom in

gap-less semiconductors.69 These spin gapless semiconductors

attract more interest because of their possible applications in

the spintronics.8,10Additional spin degree of freedom will be

a huge advantage in high-density information storage and to

control the spin polarization Furthermore, because of its high

spin polarization, it is applicable to spin filters Wang

pre-dicted that PbPdO2is a gapless semiconductor and Co-doped

PbPdO2is a spin gapless semiconductor by the first principle

calculation.8 Experimentally, it was reported that a bulk

PbPdO2 exhibits a metal-insulator-like transition and a

ferromagnetic-like behavior at low temperatures.11,12 The

metal-insulator-like transition comes from the gapless

elec-tronic band structure, and the ferromagnetic-like behavior

seems to be associated with oxygen vacancies and/or

spin-orbit coupling (SOC) More investigations on Co- and

Mn-doped PbPdO2 showed that the magnetic properties can be

easily tuned to be either ferromagnetic or antiferromagnetic.13

In a thin film form of Co-doped PbPdO2, colossal

electrore-sistance and giant magnetoreelectrore-sistance were proposed in a wide

temperature range from 25 to 300 K.14The magnetoresistance

ratio changes the sign from positive to negative when the

temperature is cooled to 27 K This sign change within such a

narrow temperature range was described by sharp phase tran-sition or sudden change of the band structure Our recent study on both PbPdO2 and Co-doped PbPdO2 thin films showed that theex-situ annealing process has a major influ-ence on the electrical transport as well as the surface mor-phology.15This infers that the oxygen in theex-situ annealing process plays an important role in determining the physical properties Especially, the magnetization of PbPdO2 shows ferromagnetic-like signal, despite there is no magnetic ion.11 The origin of the ferromagnetism seems to be oxygen vacan-cies and/or SOC Thus, we need to study the role of oxygen and the SOC mechanism by varying the annealing time

We have observed a crossover between weak anti-localization (WAL) and weak localization (WL) in Pb(Pd,Co)O2samples depending on ex-situ annealing Since the SOC plays an important role in the magnetic and electric properties of gapless and spin gapless semiconductors, the observed crossover between WAL and WL provides clues to understanding the underlying physics Such crossovers between WAL and WL have been widely observed in various systems For example, Roulleauet al.16reported the crossover

in InAs nanowires by changing the nanowire diameter, and Sch€aperset al also reported similar phenomena in GaInAs/InP nanowires where the crossover is explained by a confinement-induced effect for narrow nanowires.17Recently, such cross-overs have been found in topological insulators by changing magnetic impurity concentration, temperature, and magnetic field,5or by controlling the Fermi level with gate voltage.18 More recently, Wanget al.19have shown the crossover even in the bulk state of Bi2Se3possessing strong SOC Also in gra-phene systems, the crossover was reported due to the valley symmetry breaking in Fermi line,20 and the transition from

WL to WAL was found when the carrier density decreases and

a) Author to whom correspondence should be addressed Electronic mail:

mhjung@sogang.ac.kr

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:

the temperature increases.21Despite the variety of physical

ori-gin for the crossover, this study can call another way for better

understanding of the crossover induced by theex-situ

anneal-ing, especially in gapless semiconducting systems

The thin films of PbPdO2 and Pb(Pd,Co)O2 (herein,

PbPd0.75Co0.25O2) were deposited by a pulsed laser

deposi-tion (PLD) technique The targets for deposideposi-tion were

manu-factured by solid state reactions.11,13 The samples were

deposited on MgO (100) substrates An excimer laser of

COMPEX-102 model with a KrF(248 nm) source was used

for the PLD system A lab made optical setup with a high

vacuum chamber was used in the PLD setup The chamber

was prepared with an initial vacuum condition of 106Torr

The deposition temperature was 550C which was reached

in a rate of 10C/min The oxygen partial pressure during

the deposition was 300 mTorr The laser power was

main-tained at 0.19 W and the beam trace was fixed at a

3 mm 1 mm area for each pulse with a pulse rate of 3 Hz

After deposition the samples wereex-situ annealed in an air

with PbO powders in a box furnace for 12 h and 24 h at

650C which is theex-situ annealing temperature Note that

according to our previous study,15 the optimal annealing

conditions were 650C and 12 h The scanning electron

microscope and X-ray diffraction were used for sample

char-acterization The film thicknesses of the Pb(Pd,Co)O2 and

PbPdO2films were 60 nm and 200 nm, respectively, which

were optimal thickness with high crystallinity and smooth

surface The X-ray spectroscopy was measured to determine

the Co valence state The magneto-transport measurements

were done using the Van de Pauw method Several

differ-ence samples were made in separate annealing runs to check

reproducible data The magnetic fields were swept from9

to 9 T at low temperatures down to 2 K The physical

prop-erty measurement system (PPMS-Quantum Design) was

used to change the temperature and magnetic field

Figure 1shows the Hall resistivity qxy of as-grown and

ex-situ annealed Pb(Pd,Co)O2thin films The slopes of all the

Hall resistivity are positive, implying that the charge carriers

are mostly holes The carrier densities of the Pb(Pd,Co)O2

films are 19.02, 15.51, and 6.44 1019cm3 for the

as-grown, 12 and 24 h-annealed samples, respectively The

car-rier density decreases gradually by the increase of ex-situ

annealing time The inset of Fig.1shows the temperature de-pendence of electrical resistivity qxx The resistivity of Pb(Pd,Co)O2 is semiconducting, where the resistivity increases as the temperature decreases, i.e., dqxx/dT < 0 However, we could not fit the data with the Arrhenius law, which is lnq 1/T The mobilities are 9.33, 2.56, and 1.96 cm2 V1 s1 for as-grown, 12 and 24 h-annealed sam-ples, respectively The carrier mobility decreases by the increase of theex-situ annealing time As we reported previ-ously, the carrier density and mobility of Pb(Pd,Co)O2 are strongly influenced by Pd-O hybridization and oxygen vacan-cies onex-situ annealing.15

Now let us get to the main issue on the magneto-transport properties of Pb(Pd,Co)O2 In Fig.2, we plot the magnetoconductance (MC) curves measured at 2 K For the Pb(Pd,Co)O2case in Fig.2(a), it is hard to compare the MC curves between the as-grown and ex-situ annealed samples because of their different behaviors Here, we only focus on the low-field data between 1 and 1 T The as-grown Pb(Pd,Co)O2sample shows a sharp negative MC cusp at low fields This behavior is a finger print of WAL, which nor-mally comes from strong SOC The low-field MC curve of Pb(Pd,Co)O2shows a change from negative to positive cusp

by theex-situ annealing The annealed Pb(Pd,Co)O2samples exhibit only positive MC cusps at low fields Therefore, we speculate that the SOC is weakened byex-situ annealing and thereby the WAL is changed to WL For the PbPdO2case in Fig.2(b), the main features are the only positive MC cusp at low fields, which is different from the Pb(Pd,Co)O2 case The results are in consistent with the bulk PbPdO2 sam-ples.11,13 Since the PbPdO2 has no magnetic scattering source and shows similar positive MC behavior observed in the annealed Pb(Pd,Co)O2, the positive MC cusp in the PbPdO2 films can be interpreted as WL In that sense, we discuss the WL and WAL contributions to the MC data

In order to explain the experimental MC data of WL and WAL, the Hikami-Larkin-Nagaoka (HLN) theory has been

FIG 1 Hall resistivity q xy for as-grown, 12 and 24 h-annealed Pb(Pd,Co)O 2

thin films at 5 K The inset shows the temperature dependence of electrical

resistivity q xx for as-grown and 24 h-annealed Pb(Pd,Co)O 2 , respectively.

FIG 2 MC curves for as-grown, 12 and 24 h-annealed (a) Pb(Pd,Co)O 2 and (b) PbPdO 2 films at 2 K.

widely used in the diffusive regime of 2-dimensional (2D)

systems.22,23 Due to the layered crystal structure of

Pb(Pd,Co)O2and its 2D conduction path,24it can be

consid-ered to be a quasi-2D system although the thickness of our

films is 60 nm The original HLN theory rigorously treated

the localization problem including the spin-orbit interaction

and the magnetic scattering by impurity spins because these

interactions have different symmetries.23We tried to fit our

experimental MC data with this original HLN theory, but

could not extract any physically meaningful parameters

because of too many fitting parameters in one MC curve

Thus, we adopted the simplified empirical HLN theory with

two fitting parameters a andl/, used in many other materials

with strong spin-orbit coupling,5,25–29given by

dr Bð Þ ¼ r Bð Þ  r 0ð Þ

¼ ae

2

2ph w

1

2þ  4eBl2 U

 ln  4eBl2 U

; (1)

wheree is the electronic charge, h is the Planck’s constant,

l/is the inelastic scattering length, and w(x) is the digamma

function The a value should be either 0, 1, or1/2 These

three cases of a are distinguished by the main mechanism in

the process of electronic transport.29,30When a is 0, there is

strong magnetic scattering so that all quantum phases are

incoherent and disappear If SOC and magnetic scattering

are rather weak, WL is dominant due to the constructive

in-terference of coherent quantum phases, so that a¼ 1 If SOC

is strong but magnetic scattering is weak, a is0.5, because

WAL is the main mechanism where the destructive quantum

interference is dominant due to the strong SOC If single

mechanism is dominant, a should be one of the exact values

of0.5, 0, or 1 However, each mechanism competes each

other so that a is an intermediate value between them in

actual systems It is well known that when a is positive, the

MC curve tends to show a positive slope, and it is a

charac-teristic of commonly observed WL.5On other hand, a

nega-tive slope of MC curve is a signature of WAL, which

originates from strong SOC If a is an intermediate value, the

mechanism in the magneto-transport is in a competing

re-gime between WAL and WL.17,18,31–36

Now let us discuss the fitted results by the HLN theory

for experimental MC data of Pb(Pd,Co)O2and PbPdO2thin

films Fig.3shows the well fitted results by the HLN theory

We obtain intermediate a values for different ex-situ

annealed conditions in the Pb(Pd,Co)O2 films The fitted a

value of the as-grown Pb(Pd,Co)O2film is0.15 This

nega-tive value suggests strong SOC, which means WAL is the

dominant transport mechanism As aforementioned, the

neg-ative a (¼ 0.5) value implies strong SOC Thus, it is clear

evidence of the strong SOC in the as-grown Pb(Pd,Co)O2

film However, the a value becomes positive 0.013 and

0.0023 for the 12 and 24 h-annealed samples, respectively

The positive a implies that WL overcomes the WAL

mecha-nism In other word, the quantum interference is changed

from destructive to constructive by the ex-situ annealing

This crossover can be explained by the Co state in

Pb(Pd,Co)O2 The Co state in as-grown Pb(Pd,Co)O2is

tri-valent, which was confirmed by previous x-ray absorption

spectroscopy (XAS) experiments,37 then the trivalent Co forces the Pd state in Pb(Pd,Co)O2 to exist as Pd1þ The

Pd1þstate induced by the Co3þstate causes the Pd-O hybrid-ization to be weaker byex-situ annealing The weakened

Pd-O hybridization can result in the decrease of SPd-OC, and finally the constructive quantum interference (i.e., WL) is achieved by annealing in the Pb(Pd,Co)O2films This result

is supported by our previous report.15

On the other hand, for PbPdO2, the HLN theory may not

be suitable because of the thick film thickness of 200 nm However, we fit the MC data with a similar method to the Pb(Pd,Co)O2case for simple comparison The fitted results support the 3-dimensional (3D) transport mechanism in the PbPdO2case, which will be discussed later We obtain only positive a values (WL) for differentex-situ annealing condi-tions in the PbPdO2films The obtained a values are 0.31, 0.20, and 0.25 for the as-grown, 12 and 24 h-annealed sam-ples, respectively Since the a values are positive (closer to 1 instead of 0.5), the origin of positive a is either due to strong magnetic scattering (where a is 0) or strong WL (where a is 1) However, in the PbPdO2 films, there is no magnetic scattering source Thus, the origin of the positive

MC curves is relatively weak SOC Even though the strong SOC exists in the gapless PbPdO2films, WL can be domi-nant when the spin flip time due to the SOC is relatively lon-ger than the inelastic scattering time, or SOC scattering length is longer than inelastic scattering length l/ It means that the quantum interference between the time reversed tra-jectories are still constructive in spite of the existence of

FIG 3 Low-field MC data for as-grown, 12 and 24 h-annealed (a) Pb(Pd,Co)O 2 and (b) PbPdO 2 films at 2 K The red lines represent the fitted curves by the HLN theory As for the fitted results, we obtain a ¼ 0.15, 0.013, 0.0023 and l / ¼ 79, 297, 249 nm for the as-grown, 12 and 24 h-annealed Pb(Pd,Co)O 2 samples, respectively, and a ¼ 0.31, 0.20, 0.25 and

l / ¼ 33, 78, 74 nm for the as-grown, 12 and 24 h-annealed PbPdO 2 samples, respectively.

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:

SOC Similar phenomena were already reported in

topologi-cal insulators by Wanget al.19 This is also consistent with

the results of annealed Pb(Pd,Co)O2 The opposite annealing

effect in PbPdO2, compared to Pb(Pd,Co)O2, can be

explained by oxygen vacancies, which influence the Pd-O

hybridization.29,30As the samples are annealed, the oxygen

vacancies are reduced and the Pd1þstates are reduced, unlike

the Pb(Pd,Co)O2 This causes the Pd-O hybridization to be

enhanced, which well agrees with the results of XAS and

photo emission spectroscopy (PES) experiments.29 Also, in

our previous paper,15 the carrier density of PbPdO2 is

decreased after the ex-situ annealing and the mobility is

increased due to the enhanced Pd-O hybridization Then, the

enhanced Pd-O hybridization can lead to the increase of

SOC, and finally the a value decreases with the ex-situ

annealing in PbPdO2

By fitting the MC data with the HLN theory, we can

obtain another important parameter, the inelastic scattering

lengthl/ If the inelastic scattering length is longer than the

thickness of the sample, the transport mechanism can be

treated by a 2D regime.22 For the as-grown, 12 and 24

h-annealed Pb(Pd,Co)O2films, the inelastic scattering lengths

are 79, 297, and 249 nm, respectively When we consider the

thickness (¼60 nm) of the Pb(Pd,Co)O2film, it implies that

the transport mechanism is in the 2D regime because thel/

is longer than the sample thickness This result is also

rea-sonable when considering the relation between l/ and a.36

However, for the as-grown PbPdO2 films, the l/ is 33 nm

which is shorter than the sample thickness This hints that

the transport mechanism is in the 3D regime, as

aforemen-tioned, where many scattering events occur in the

out-of-plane direction of the film However, by annealing for 12

and 24 h, thel/increases to 78 nm and 74 nm, respectively

This increment ofl/ suggests that the transport mechanism

approaches the 2D transport regime Considering the relation

betweenl/and a, this result is also reasonable

In order to analyze the WAL behavior (i.e., negative

MC curve) of the as-grown Pb(Pd,Co)O2film more closely,

the temperature-dependent data of low-field MC have been

measured as plotted in Figs.4(a)and4(b) It shows that the

sharp negative cusp in MC gradually broadens as the

temper-ature increases up to 5 K, while it changes to a broad positive

cusp above 5 K The fittedl/decreases gradually as the

tem-perature increases from 2 K to 9 K The l/ is 80 nm at 2 K

and monotonically decreases to 20 nm at 8 K, as shown in

Fig.5 Here, it is noticeable that thel/at 5 K is 50 nm that is

near the thickness of Pb(Pd,Co)O2film This infers that the

transport mechanism can be changed around 5 K Indeed, the

a value changes from negative to positive with increasing

temperature at 5 K, as marked with green horizontal line in

Fig.5 As aforementioned, thel/is closely associated with

the a The negative a means strong SOC below 5 K, where

we have largerl/values than the sample thickness, and vice

versa Similar crossover due to the length scale has been

reported in GaInAs/InP nanowire.17 They reported that the

crossover between WAL and WL occurs when the spin

relaxation length is suppressed by reducing the diameter of

nanowires

Furthermore, the temperature dependence of l/is

well-known to be scaled asl/ Tp/2.25In Fig 5, we plot thel/

data as a function of temperature in logarithmic scales The data are well fitted with p¼ 1.1 6 0.16 when T < 5 K and

p¼ 2.2 6 0.44 when T > 5 K The p value determines the col-lision type of inelastic scattering When the electron-electron collision is dominant,p¼ 1, while the electron-phonon colli-sion is dominant, p¼ 3 Our experimental data suggest that when temperature is below 5 K, the phonon scattering is sup-pressed and electron-electron scattering prevails For higher temperature, electron-phonon scattering starts to contribute

In our observation, when temperature is raised above 5 K, three important changes occur simultaneously: (1) the MC mechanism (sign of a) is changed from WAL to WL, (2) the 2D nature is weakened (l/< film thickness), and (3) the pho-non scattering become important (p 2.2) We speculate that those three changes are mutually related each other

We observed WAL signals from the strong SOC in Pb(Pd,Co)O2films By ex-situ annealing, the SOC is weak-ened and the crossover from WAL to WL is observed These results are compared with WL signals for PbPdO2 The dif-ferent magneto-transport mechanism was interpreted in terms of a and l/ parameters based on the HLN theory, which are closely associated with the Pd-O hybridization of

FIG 4 (a) Low-field MC data of as-grown Pb(Pd,Co)O 2 film measured at various temperatures 2, 3, 4, 5, 8, and 9 K The red lines represent the fitted curves by the HLN theory (b) Corresponding magnetoresistance (MR) data measured at various temperatures 2, 3, 4, 5, 7, 8, and 9 K.

FIG 5 Fitted l / and a values as a function of temperature for as-grown Pb(Pd,Co)O 2 film The red and blue dashed lines are the cases with p ¼ 1 and 3, respectively The red and blue solid lines represent the linear fits from the experimental data for two temperature regimes, T < 5 K and T > 5 K, respectively The green horizontal line indicates zero a.

the two films This strong dependence of SOC associated

with Pd-O hybridization by annealing process shows the

potential of Pb(Pd,Co)O2as a spintronics material

This work was supported by the National Research

Foundation of Korea (NRF) (Nos 2014R1A2A1A11050401,

2013R1A1A2011936, and 2012M2A2A6004261)

1

I M Tsidilkovski, Electron Spectrum of Gapless Semiconductors

(Springer, New York, 1996).

2 X L Wang, S X Dou, and C Zhang, NPG Asia Mater 2, 31 (2010).

3 Y Li, Z Zhou, P Shen, and Z Chen, ACS Nano 3, 1952 (2009).

4

M I D’yakonov and A V Khaetskl, Zh Eksp Teor Fiz 82, 1584 (1982)

[Sov Phys JETP 55, 917 (1982)].

5 M Liu, J Zhang, C.-Z Chang, Z Zhang, X Feng, K Li, K He, L.-L.

Wang, X Chen, X Dai, Z Fang, Q.-K Xue, X Ma, and Y Wang, Phys.

Rev Lett 108, 036805 (2012).

6

Y Pan and Z Yang, Phys Rev B 82, 195308 (2010).

7 S Ouardi, G H Fecher, and C Felser, Phys Rev Lett 110, 100401 (2013).

8 X L Wang, Phys Rev Lett 100, 156404 (2008).

9

G Z Xu, E K Liu, Y Du, G J Li, G D Liu, W H Wang, and G H.

Wu, EPL 102, 17007 (2013).

10 S A Wolf, D D Awschalom, R A Buhrman, J M Daughton, S von

Molnar, M L Roukes, A Y Chtchelkanova, and D M Treger, Science

294, 1488 (2001).

11

K J Lee, S M Choo, J B Yoon, K M Song, Y Saiga, C.-Y You, N.

Hur, S I Lee, T Takabatake, and M H Jung, J Appl Phys 107, 09C306

(2010).

12

T C Ozawa, T Taniguchi, Y Nagata, Y Noro, T Naka, and A.

Matsushita, J Alloys Compd 388, 1 (2005).

13 K J Lee, S M Choo, Y Saiga, T Takabatake, and M.-H Jung, J Appl.

Phys 109, 07C316 (2011).

14

X Wang, G Peleckis, C Zhang, H Kimura, and S Dou, Adv Mater 21,

2196 (2009).

15 S M Choo, K J Lee, S M Park, G S Park, and M.-H Jung, J Appl.

Phys 113, 014904 (2013).

16

P Roulleau, T Choi, S Riedi, T Heinzel, I Shorubalko, T Ihn, and K.

Ensslin, Phys Rev B 81, 155449 (2010).

17 Th Sch€ apers, V A Guzenko, M G Pala, U Z€ ulicke, M Governale, J Knobbe, and H Hardtdegen, Phys Rev B 74, 081301(R) (2006).

18

M Lang, L He, X Kou, P Upadhyaya, Y Fan, H Chu, Y Jiang, J H Bardarson, W Jiang, E S Choi, Y Wang, N.-C Yeh, J Moore, and K L Wang, Nano Lett 13, 48 (2013).

19

H Wang, H Liu, C.-Z Chang, H Zuo, Y Zhao, Y Sun, Z Xia, K He, X.

Ma, X C Xie, Q.-K Xue, and J Wang, Sci Rep 4, 5817 (2014).

20 E McCann, K Kechedzhi, V I Fal’ko, H Suzuura, T Ando, and B L Altshuler, Phys Rev Lett 97, 146805 (2006).

21

F V Tikhonenko, A A Kozikov, A K Savchenko, and R V Gorbachev,

Phys Rev Lett 103, 226801 (2009).

22 P A Lee and T V Ramakrishnan, Rev Mod Phys 57, 287 (1985).

23

S Hikami, A I Larkin, and Y Nagaoka, Prog Theor Phys 63, 707 (1980).

24 S W Chen, S C Huang, G Y Guo, J M Lee, S Chiang, W C Chen,

Y C Liang, K T Lu, and J M Chen, Appl Phys Lett 99, 012103 (2011).

25

R Scherwitzl, S Gariglio, M Gabay, P Zubko, M Gibert, and J.-M Triscone, Phys Rev Lett 106, 246403 (2011).

26

J Kim, S Lee, Y M Brovman, M Kim, P Kim, and W Lee, Appl Phys Lett 104, 043105 (2014).

27 J Hu, J Y Liu, and Z Q Mao, J Phys.: Condens Matter 26, 095801 (2014).

28 S.-P Chiu and J.-J Lin, Phys Rev B 87, 035122 (2013).

29 T He, G Wang, T Zhang, I.-K Sou, G K L Wong, J.-N Wang,

H.-Z Lu, S.-Q Shen, and F.-C Zhang, Phys Rev Lett 106, 166805 (2011).

30 J A Kurzman, M.-S Miao, and R Seshadri, J Phys.: Condens Matter 23,

465501 (2011).

31

B Y Jin and J B Ketterson, Phys Rev B 33, 8797(R) (1986).

32 S A Studenikin, P T Coleridge, N Ahmed, P J Poole, and A Sachrajda, Phys Rev B 68, 035317 (2003).

33

I Garate and L Glazman, Phys Rev B 86, 035422 (2012).

34 H Suzuura and T Ando, Phys Rev Lett 89, 266603 (2002).

35 J B Miller, D M Zumb€ uhl, C M Marcus, Y B Lyanda-Geller, D Goldhaber-Gordon, K Campman, and A C Gossard, Phys Rev Lett 90,

076807 (2003).

36 H.-Z Lu, J Shi, and S.-Q Shen, Phys Rev Lett 107, 076801 (2011).

37 D H Kim, J H Hwang, E S Lee, K J Lee, S M Choo, M H Jung, J Baik, H J Shin, B J Kim, K Kim, B I Min, and J.-S Kang, Appl Phys Lett 104, 022411 (2014).

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: